Skip to main content

Full text of "Journal of ethnobiology."

See other formats


Journal and Society Organization 



EDITOR: Deborah M 



Missouri. Columbia, MO 



ASSOCIATE EDITOR 



Mexico 



NEWS & COMMENTS EDITOR: Gary J. Martin, 94 Blvd. Flandrin, 75116, Paris 



<: 33/1/45533001. 
REVIEW EDITOR 



Cruz 



Bulhoes-Manguinhos, 21.041 Rio de Janiero-RJ-BRASIL. 
BOOK REVIEW EDITOR: Nancy J. Turner, Environmental Studies Program, University 

of Victoria, Victoria, B.C. V8X 3P4 CANADA . 
EDITORIAL ASSISTANT: Christopher B. Pulliam, American Archaeology Division, 

103 Swallow Hall, University of Missouri, Columbia, MO 65211. 
PRESIDENT: Paul Minnis, Department of Anthropology, University of Oklahoma, 

Norman, Oklahoma 73019. 
PRESIDENT-ELECT: Cecil H. Brown, Department of Anthropology, Northern Illinois 

University, DeKalb, Illinois 60115. 
SECRETARY/TREASURER: Catherine S. Fowler, Department of Anthropology, University 

of Nevada, Reno, Nevada 89557. 
CONFERENCE COORDINATOR: Jan Timbrook, Department of Anthropology, Santa 

Barbara Museum of Natural Historv, 2559 Puesta Del Sol Road, Santa Barbara, 



California 93105. 



BOARD OF TRUSTEES 



ROBERT A. BYE, JR., Universidad Nacional Autonoma de Mexico, MEXICO 

botany, ethnoecology . 
EDELMIRA LINARES. Universidad Nacional Autonoma de Mexico, MEXICO 



botany. 



J. REITZ, University 



Ex officio: Past Presidents Steven A. Weber, Amadeo M. Rea and Elizabeth S. Wing 
Permanent board member Steven D. Emslie; The Editor, President, President Elect 
Secretary/Treasurer, and Conference Coordinator. 

EDITORIAL BOARD 

KAREN R. ADAMS, Crow Canyon Archaeological Center, USA; paleoethnobotany . 
JANIS B. ALCORN, World Wildlife Fund, Washington, DC, USA; ethnobotany, tradi 

tional agriculture, ethnomedicine . 
BRENT BERLIN, University of California, Berkeley, USA; ethnobiological classification 

medical ethnobotany. 



systems 



BROWN, Northern Illinois University, DeKalb, USA; folk biological classification 
HARRIS, University College, London, ENGLAND; ethnoecology, subsistena 



JOHNS, McGill University, CANADA; chemical ecology, 



HARRIET V. KUHNLEIN, McGill University, CANADA; ethnonutrition , human m 
GARY J. MARTIN, Grupo de Apoyo al Desarrollo Etnico, Oaxaca, MEXICO 



4 * ■ 



classification 



Instituto 
ethnoecology, ethnoentomology , tropical cultural ecology 



natural 



AMADEO M. REA, San Diego Natural History Museum, USA; cultural ecology, zooarcha- 

eology, ethnotaxonomies . 
ELIZABETH J. REITZ, University of Georgia, USA; zooarchaeology. 
MOLLIE S. TOLL, University of New Mexico, USA; prehistoric and historic ethnobotany. 
WILLARD VAN ASDALL, Past Editor, Tucson, Arizona, USA; ethnobiology . 

Feature editors Carlos E.A. Coimbra and Nancy J. Turner (see above). 

Journal of Ethnobiology is published semi-annually. Manuscripts for publication, information for the "News and 
Comments" and book review sections should be sent to the appropriate editor on the inside back cover of this issue. 



©Society of Ethnobiology 
ISSN 0278-0771 












Journal of 

Ethnobiology 



Missouri Barmen 



APR 1 1993 



GARDEN LIBRARY 



VOLUME 12, NUMBER 1 SUMMER 1992 



SIXTEENTH ANNUAL 
ETHNOBIOLOGY CONFERENCE 



11-13 MARCH 1993 



BOSTON UNIVERSITY 
BOSTON, MASSACHUSETTS / U.S. A 



CONTACT 

Dr. Julie Hansen 

Department of Archaeology 

675 Commonwealth Avenue / Room 347 

Boston University 

Boston, MA 02215 

Telephone: (617) 353-3415 









CONTENTS 



EDITOR'S VIEW 



1 



EDIBLE WOOD FERN ROOTSTOCKS OF WESTERN NORTH 
AMERICA: SOLVING AN ETHNOBOTANICAL PUZZLE 

Nancy J. Turner, Leslie M. Johnson Gottesfeld, 

Harriet V. Kuhnlein, Adolf Ceska 



1 



INFLUENCES OF MID-HOLOCENE ALTITHERMAL CLIMATES 
ON MAMMALIAN FAUNAS AND HUMAN SUBSISTENCE 
IN EASTERN WASHINGTON 

R. Lee Lyman 



37 



AN OPTIMAL FORAGING ANALYSIS OF PREHISTORIC 
SHELLFISH COLLECTING ON SAN CLEMENTE ISLAND, 

CALIFORNIA 

L. Mark Raab 



63 



TWO PREHISTORIC PUEBLOAN AVIFAUNAS FROM THE 
PECOS VALLEY, SOUTHEASTERN NEW MEXICO 

Steven D. Emslie, John D. Speth, Regge N. Wiseman 



83 



FOOD TABOOS AT BUZIOS ISLAND (BRAZIL): THEIR 
SIGNIFICANCE AND RELATION TO FOLK MEDICINE 

Alpina Begossi 



117 



SHORT COMMUNICATIONS 

Eduardo O. Kohn 141 

Leslie M. Johnson Gottesfeld 153 



NEWS 157 



BOOK REVIEWS 34, 62, 81, 139, 156 




For the past year, the Society of Ethnobiology has been actively recruiting 
new members. I'd like to extend a special welcome to all those who are reading 
their first issue of the Journal. It seems a good time to say a few words about how 
the papers you will be reading came to be published. 

The Journal publishes three types of refereed manuscripts: articles, short 
communications, and comments. An article describes original research in any area 
of ethnobiology. Length is variable: you must present enough background infor- 
mation, detail on methodology, and basic data for an educated reader to evaluate 
the validity of the conclusions you have drawn, while avoiding wordiness, repeti- 
tion, and jargon. English, Spanish, and French abstracts are required for articles. 
A short communication is a brief presentation of the results of limited or 
preliminary research. No abstracts are required. A comment is a discussion of 
an article or short communication previously published in the Journal. Like 
articles and short communications, all comments are sent out for peer review. 
If the comment is accepted for publication, the author of the original article is 
invited to write a response and comment and response are published together. 

After a manuscript is submitted (original and two copies), the editor sends 
it to an editorial board member, who chooses two reviewers. (Send me your name, 
address, phone/FAX/EMail numbers, and a description of your expertise in 
ethnobiology if you would like to review manuscripts.) After all reviews are back, 
a process that takes 8-12 weeks, the editor makes the final decision and informs 
the author that the manuscript (1) is accepted, (2) is accepted with minor revi- 
sions, (3) should be revised and resubmitted, or (4) is rejected. A great many 

' stage and are ultimately 



manuscripts go through the "revise and resubmit" stage and 

published: revise and resubmit is not rejection. The final stages of the process 

are copyediting, typesetting, and proofing, proofing, and more proofing . 

The style guide for the Journal is published in issue 10(2); you may also write 
me for a copy. A few reminders for new and experienced submitters alike: 

French, Spanish, and English abstracts are required for article- length submis- 



ing your original abstract is a real problem, we u nei 
simDlv not the time or the money for the Journal staff 



everyone 



aced means doubled spaced everywhere in the manuscript: al 
text, quotations, tables, bibliography, notes, and figure captions. Re 
often write comments on the manuscript; the editor must copyedit 



i 



typesetter. These tasks are made very difficult if sections of the text are n 

doubled spaced. 

Word processors are wonderful, but resist the temptation to use fancy type 

underline species and genus names and foreign words (no italic type, please); 

bold face may be used for foreign words for special emphasis (e.g., in an 

English text with both Spanish and Quechua common names, you may use 

bold face for Quechua), but is not required. 

You must provide page numbers for all quotations, long or short, and in an 

English text translations of any quotes not in English. 

Double check all scientific and foreign words for accuracy of spelling and 

accenting— the editors cannot do this for you. 

Double check the bibliography: most queries to authors during the copyediting 

process have to do with missing first names for authors and editors, missing 

publication locations, and the spelling out of abbreviations. 

My thanks to all the board members and reviewers who assisted with manu- 
scripts for issue 12(1). And suecial thanks to Chris Pulliam. who filled in for two 



this job. 



interim News and Comments editor. Gary Martin now 



DMP 



/. Ethnobiol. 12(1): 1-34 



Summer 1992 



EDIBLE WOOD FERN ROOTSTOCKS OF WESTERN NORTH 
AMERICA: SOLVING AN ETHNO BOTANICAL PUZZLE 






NANCY J. TURNER 
Environmental Studies Program 

University of Victoria 
Victoria, B. C V8W 2Y2 

LESLIE M. JOHNSON GOTTESFELD 

Department of Anthropology 

University of Alberta 
Edmonton, Alberta T6G 2H4 

HARRIET V. KUHNLEIN 

School of Dietetics and Human Nutrition 

Macdonald College of McGill University 

Ste. Anne de Bellevue, P.Q. H9X ICO 

ADOLF CESKA 

Royal British Columbia Museum 

Victoria, B.C. V8V 1X4 



ABSTRACT.— Many ethnographic reports refer to a large, "pineapple-like" fern 
rootstock which was an important native root vegetable in northwestern North 
America. It is suggested here that the primary, most commonly used edible type 
is Dryopteris expansa, with other, related species having been used in some localities 
and under some circumstances. The rootstocks were cooked in pits, often in winter 
when food was scarce. They are seldom eaten today and are known primarily 
by native elders. Several botanical identifications for this food have been sug- 
gested in Dryopteris, Athyrium, and other fern genera. Species verification has 
been complicated by lack of botanical expertise among early reporters, difficulty 
in recalling the fern characters by elders, and taxonomic complexities of the ferns. 
Rootstocks of D. expansa were harvested by the Nuxalk at Bella Coola, cooked, 
and prepared for nutrient analyses. Proximate composition and energy are similar 
to that of the common potato, but Ca, Mg, Zn, Cu, and Mn were present in levels 
several-fold higher than that of potato. 



RESUMEN.— Varios informes etnograficos se refieren a un rizoma grande, "seme- 
jante a una pina," de un helecho, que era una importante raiz alimenticia nativa 
en el noroeste de Norteamerica . Se sugiere aquf que el tipo comestible primario, 
mas comunmente usado, es Dryopteris expansa, habiendose empleado otras especies 
emparentadas en algunas localidades y en ciertas circunstancias. Los rizomas se 
cocian en cavidades en el suelo, con frecuencia en el invierno cuando escaseaba 
la comida. En la actualidad se comen muy poco, y son principalmente los ancianos 
indigenas quienes los conocen. Para este alimento se han sugerido varias identi- 
ficaciones botanicas en Dryopteris, Athyrium y otros ge'neros de helechos. La 
verification de la especie se ha complicado por la falta de pericia botanica entre 
los orimeros observadores, la dificultad en recordar las caractensticas de los 



2 



TURNER et al. Vo1 - 12 ' No - X 



parte de los ancianos, y la complejidad taxonomica 



nzomas 



nutricional 
arrojaron resultados similares a los de la papa comun 



energetico, pero se encontraron niveles de calcio, magnesio, cine, cobre, y 
manganeso varias veces mas altos que los de la papa. 

RESUME.-De nombreux rapports ethnographiques font reference a de grosses 
racines de fougere, semblables a un ananas, qui etaient un important legume local 
en Amerique du Nord. L'hypothese est emise que le type comestible pnmaire 
et le plus utilise est la Dryopteris expansa alors que d'autres especes proches etaient 
utilisees dans certaines localite's et dans certaines circonstances . Les racmes etaient 



nournture 



consommees 



ations botaniques pour cet aliment ont ete sug- 
ge^Vdans Dryopteris, Athyrium et d'autres genres de fougeres. La verification 
botanique a ete rendue difficile par le manque d' expertise botanique des premieres 
sources, par la mauvaise memoire des anciens a propos des caracteristiques des 
fougeres, et par leur complexite' taxonomique . Les racines de D. expansa furent 
recoltees par les Nuxalk a Bella Coola, cuites et preparers pour etre analysees. 
La composition "proximate" et l'energie sont similaires a celles de la pomme de 
terre commune mais le Ca, Mg, Zn, Cu, et Mn y sont presents en plus grande 
quantite'. 



INTRODUCTION 



Ferns are an important component of northwestern North American flora. 
The moist, mild coastal climate is ideal for the growth of many Pteridophyte 
species, and the diversity of habitats and microclimates is reflected in the diver- 
sity of species of ferns . Many reports have been published concerning the edibility 
of certain ferns by native peoples of the region (cf . Gunther 1973; Turner 1975). 
The use of the rhizomes of western bracken fern (Pteridium aquilinum (L.) Kuhn) 
as food by Northwest Coast peoples has been well documented (cf . Norton 1979) . 
The rhizomes of licorice fern (Polypodium glycyrrhiza D.C. Eat.) are also well known, 
even today, as a mouth sweetener and appetizer, although they were seldom 
eaten in any quantity (cf. Turner and Bell 1973; Turner 1973, 1975). 

Perhaps the most intriguing and puzzling "fern food" is the rootstock 
recalled by many contemporary native elders and widely cited in the literature 
as a large, "pineapple-like," or "banana-like" clump of "fingers" which was 
cooked and eaten traditionally, both as a regular part of the diet and as emer- 
gency rations, along the coast from Oregon to Alaska and inland in British 
Columbia. There is probably more confusion about the identity of this edible fern 
rootstock, hereinafter called wood fern in a general context, than about any other 
traditional food plant in northwestern North America. 

One of the earliest literature references to edible wood fern is by Gorman 
(18%), who identified it as Dryopteris spinulosa ["Aspidiutn spinulosum var. dilatatum 
(Wood-fern)"] and provided the following account of its use by the Coas 
Tsimshian of southeastern Alaska (Gorman 1896:78-79): 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 3 



[Wood-fern] ... is abundant in rich open woods near sea level, and the 
rootstock or caudex is highly relished by the natives, who cook and eat 
it in large quantities, it being the first vegetable food which they obtain 
in early spring. [It] ... is called "Ahh, " by the Tsimsians, who inform 
me that it is larger, sweeter, and of better flavor when grown under or 
in the vicinity of salmon berry bushes . . . 



Ethnographers without botanical knowledge described the fern in the mos 
general of terms. For example, J.P. Harrington, in his unpublished ethnographi 
notes, referred to it under a Suquamish place name for a creek at Miller Ba; 
("Miller's Inlet") on the Port Madison Indian Reservation in Washington: 



du ts'kweb Means:- ts'ak 
marshy place. Have heavy 



in 



bunches. 1 



white inside. Grow up like finger 



Barbara Lane (personal communication to NJT, 1984) described the same 
Suquamish fern in more detail, based on descriptions by Native people she 

consulted: 



It [ts'Ekwi] is styled variously the "fossil fern," "evergreen fern,' 
"Indian banana." It is a tall plant, and grows on logs in damp pi 
A cluster of edible nods is found at the bottom of the stalks, lookinj 



lands placed palm to palm. The Indians gathered 
them as they did clams, burying them in a pit v 



The 



name . . . Miller Bay used to be called Squaib Bay, clearly a rendering 



term 



Edible wood fern rootstocks, all having more-or-less similar descriptions 
id methods of gathering and preparation, have been described and identified 
Dryopteris austriaca (Turner 1975; Heller 1976; Oswalt 1957), D. filix-mas (Turner 



an 



[sic]), D.expansa 



al. 1985), D. spinulosa (Gorman 1896-" Aspidium spinulosum var. dilatatum;" Harlan 
Smith, unpublished notes, 1920-19212— "Aspidium spinulosum"), Thelypteris limbo- 
sperma (Norton 1981), Cystopteris bulbifera ('Ksan, People of 1980), Athyrium fit 



femina (Kari 1987; Harlan Smith 
sorum") 



cyclo- 
Nancy 



Turner unpublished notes on Haida, 1971). 3 D. spinulosa, D. dilatata, D. austriaca, 
and D. expansa are considered to be in the same taxonomic complex, as explained 
later. In most, if not all cases, the identifications were made on the basis of selec- 



information provided 



some 



from confused or mistaken identifications. In this paper 



wood fern rootstock— the one normally 



4 



TURNER et al. Vol. 12, No. 1 



- 

Dryopteris expansa, bi; 
were also consumed 



similar types 



The confusion surrounding the identification of the edible wood fern is 
exemplified in the unpublished notes of Harlan I. Smith on plant use by the 
Nuxalk (Bella Coola) and Carrier Indians of central British Columbia. Smith refer- 
red in several places to an edible fern called sq'^alm ("squalum"), which had 
a " . . . part like a pineapple at the base [that was] used for food." In his Carrier 
ethnobotanv notes, he identified it as Athyrium filix-femina* and noted, "Leaves 



bifurcated 



Bww „, , # n as Dryopteris expansa 5 and also attributes the name 

"Xala" to this species. Then, he notes: "Trouble. Same species given two dif- 
ferent Bellacoola names, one useful for food only, the other used for medicine." 
His native Bella Coola consultant evidently applied the two native terms to dif- 
ferent specimens of the same species. Adding further confusion, a third term, 
"kamatz," was applied by Smith's Native consultants to the tops (fronds) of 
Dryopteris expansa (see previous footnote) and Athyrium filix-femina* with a note 
under the latter: "Trouble. Joshua [Moody] calls this the same as Shield Fern 
7 (54M) with which it was collected; yet that is the same species as 7 (77M) which 
he calls squalum [both of these are actually Dryopteris expansa J." 

Elsewhere in his unpublished notes, Smith mentioned two other fern species 
which were sometimes mistaken for the real squalum. Under Polystichum 
sword fern, he noted: "... This plant was not used for food . . . 
[sometimes] mistaken ... for another fern called squalum. The two forms were 
distinguished by the leaves as the banana-like parts appear the same and both 
are easy to pull up. The leaves of the sword fern ... do not die in winter . . . 



but 



unidentified fern "See el 4 tana," 7 he noted that this one was 

a little different from squalum. Sometimes people . . . mistake it for squalum. 
It is not good food. Squalum is [a good food] . . . Roots same and top same as 
squalum. I think squalum has no spores and See el i tana has." Smith's Carrier 

notes are equally confusing. 8 

As will be seen, research within the last two decades at Bella Coola has resulted 
in clarification of some of the confusion surrounding sq'walm, but the problems 
that faced Harlan Smith and the Native people he worked with seem to have 
been repeated in virtually every area where wood fern was eaten. Furthermore, 
plant taxonomists have also been confounded in the scientific classification of 
wood ferns, and the many synonyms and nomenclatural discrepancies for various 
Dryopteris species have made the problem of identification especially difficult. 

The aim of this paper is to resolve, as much as possible, the questions arising 
from the bewildering and sometimes conflicting array of botanical and 
ethnobotanical information on the edible wood fern in northwestern North 
America, and to document the traditional importance of this food in Native 
cultures. We will summarize the botanical and taxonomic features of the wood 
fern complex in northwestern North America, and relate the botanical 
characteristics to the utility of the plant. Then we will review the Native termi- 
nology, descriptions, and use of the edible wood fern, using information derived 
both from literature sources and from interviews with contemporary Native 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 5 



consultants in various speech communities of the region. Nutrient data from 



specimens 



importance of wood fern in mythology 



Information 









BOTANICAL AND TAXONOMIC ASPECTS 

OF THE WOOD FERN PUZZLE 

Narrowing down the species.— The fern species reported in this study to have 
edible, ' 'pineapple-like' ' rootstocks fall into five genera. These are annotated and 
evaluated as follows (see also descriptions in Scoggan 1978 and Taylor and 
MacBryde 1977): 

Cystopteris — C. bulbifera ('Ksan, People of 1980); does not occur in British 
Columbia (Scoggan 1978); concluded to be erroneous identification; 

Thelypteris — T. limbosperma (Norton 1981); voucher specimen (613 WTU) 
examined by Adolf Ceska in 1989 and found to be Athyrium filix-femina; 

Athyrium — only A. filix-femina (L.) Roth (lady fern) has rootstocks large 
enough and distribution wide enough to "qualify;" has been identified on occa- 
sion as the edible type (see previous entry), but usually rejected; 

Polystichum — only P. munitum (Kaulf .) Presl (sword fern) has large enough 
rootstocks to be the edible type, and has occasionally been suggested as such 
in literature citations, 10 P. braunii (Spenner) Fee (Braun's holly fern) has rootstocks 
large enough, but has never been cited in ethnobotanical or ethnographic literature 
as having been eaten; 11 

Dryopteris — most frequently cited as having edible rootstocks; taxonomic 
treatments for the genus widely variable (cf. Hulten 1968; Calder and Taylor 1968; 
Scoggan 1978); some species ruled out because of size or range, 12 leaving D. filix- 
mas (L.) Schott (male fern) and two species of the "Dryopteris expansa complex 
D. carthusiana (Villars) H.P. Fuchs^ and D. expansa (K.B. Presl) Fraser-Jenkins 
and Jermy; (see Walker 1955, 1961; Britton 1962, 1968, 1972; and Widen and 
Britton 1971 for detailed discussions of the taxonomy of this complex). 

Of the Dryopteris species mentioned, D. filix-mas was identified in one case 
(Turner 1973) but the Dryopteris expansa complex has been more consistently 
recognized. The two species from this complex are similar, but D. carthusiana (syn. 
D. spinulosa [O.F. Muller] Watt) has fronds with more or less parallel sides at their 
lower part and has a chromosome count of 2n = 164, whereas D. expansa (syn. 
D. assimilis S. Walker, D. dilatata [Hoffm.] A. Gray var. alpina Moore) has fronds 
more or less triangular in outline, with a chromosome count of 2n = 82. The 
latter is common in northwestern North America; the former is less frequent, but 
where it occurs, it may grow side by side with D. expansa (e.g., in Exchamsiks 
River Provincial Park in northern British Columbia). Of all the species men- 
tioned, the one most frequently and consistently cited as THE edible wood fern 
of northwestern Indigenous Peoples, when synonymy is considered, is Dryopteris 
expansa. 



ir 



Native perceptions. —Given the complex situation of so many, widely variable 
species and genera sharing at least some characteristics, it is not surprising 



6 



TURNER et al. Vol. 12, No. 1 



Native people have often provided varying and confusing identifications and 
descriptions of the edible wood fern and others they consider to be related. 

McHwraith (1948, in his Appendix C), in discussing folk classification of the 
Nuxalk (Bella Coola) adds another perspective to the problem: "In regard to 
plants, a difficulty lies in the fact that Bella Coola nomenclature is not always 
strictly botanical. Two or more distinct ferns, for example, may be grouped 
together on account of their similar use as food and one name applied to them 
indiscriminantly ..." A Haida Native elder, the late George Young, stated the 
same concept in discussing several ferns given the same Haida names 14 : "One 
word can mean so many names. It's all in the way the sentence or the subject 
is brought up. You can have one word for two completely different kinds of 

plants ..." 

The close perceptual relationship among various ferns is shown in the Nuxalk 
area by the term qaxmats, mentioned previously in the introductory discussion 
of Harlan Smith's unpublished notes, which is sometimes applied to the fronds 
of Athyrium, Dryopteris, Polystichum, and Pteridium 15 although the rhizomes are 
usually given different names. The identification of the true edible wood fern, 
sq'Walm, must be made from the appearance of the rootstock, not the aerial 
parts. Nuxalk elder Felicity Walkus once applied the name to specimens of both 
Dryopteris expansa and Athyrium filix-femina growing side by side 16 , but when she 
examined the rootstocks after they were dug up, she stated that only the Dryopteris 
was the true sq' w altn. This identification was confirmed by Margaret Si Wallace 
and two other Nuxalk elders. The Athyrium rootstock, they said, was the one 
called xala, which was like that of the edible fern, but smaller, blacker, and covered 
with dark hairs. Grizzlyibears are said to like to esAxala, but it makes them cranky 
(Nancy J. Turner, unpublished notes on Nuxalk, 1981). Margaret Siwallace later 
stated that there were four types of fern that were almost alike. The one with 
fleshy, round "fingers" (leaf bases) was sq'Walm, the "flat one," xala, was 
poisonous, and two other kinds are similar to sq'wdlm, and edible, but not quite 
as good (Nancy J. Turner, unpublished notes on Nuxalk, 1983). 

Lillooet consultants definitely excluded Athyrium as the edible wood fern, 
although one elder called it "a kind oi"c'?k w a? (Dryopteris) . Its big, black "roots" 
are said to be so tough that even a plough cannot go through them (Randy 
Bouchard, personal communication to NJT, 1974). One specimen collected with 
Lillooet elders and said to be the edible kind was identified as D. carthusiana 17 

In the Skeena drainage of northern British Columbia, a number of different 
identifications have been made of ax, the plant with edible rootstocks. Athyrium 
filix-femina and Dryopteris filix-mas plants in leaf were pointed out as ax by David 
Green, a knowledgeable Gitksan elder (Leslie M.J. Gottesfeld, unpublished notes, 
1988). 18 Later in the season, however, he identified Dryopteris expansa as ax, and 
excluded Athyrium. Another Gitksan consultant, Jeff Harris Sr. of Kispiox, who 
as a child had frequently collected edible fern rootstocks with his grandmother, 
pointed out that the edible type was recognized from the rootstock, rather than 
from the tops. The fern was sought and dug up after the fronds had withered. 
Ferns with a small rootstock, or ones with small or flattened leaf bases, were 
called dumtx, a term signifying "non-edible fern." During a field trip specifi- 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 7 



cally undertaken to identify and collect ax, plants of Athyrium filix-femina and 
Dryopteris filix-mas, and a small specimen of D. expansa were all classified as dumtx 
by Jeff Harris. The last was described as being almost like ax, but too small. Its 
broken "fingers" or leaf bases had the distinctive pea green colour of ax. It was 
collected in a small organic soil pocket in the hemlock forest, not in the snowbed 
or avalanche chute habitat preferred by the Gitksan in their interior geographic 
location. Larger, more robust specimens of D. expansa, found in the snowbed 
locality, were dug up and pronounced to be ax, because of the relatively large 
size of the rootstock, the diameter of the "fingers," and their green color (Leslie 
MJ. Gottesfeld, unpublished notes, 1988. 19 

Additional insight into native perception and classification of ferns comes from 
Gordon Robinson, a Haisla elder from Kitamaat Village. When shown a specimen 
oi Dryopteris expansa and asked if it was the edible fern root, he commented, "You 
can tell if it is the right fern if the 'fingers' are round [in cross section], pale green 
inside, and brittle. It grows in the forest on top of fallen logs and on stumps. 
It grows at the base of slide areas . . . You can kick it out of the ground or pull 
it out easily. Other ferns, you need a pick [to dig up] . . . "(Leslie M.J. Gottesfeld, 
unpublished notes, 1988). 20 It is significant that when one collects a fresh 
rootstock of Athyrium filix-femina, the "fingers" are tan inside, woody textured, 
and triangular in cross-section. In addition, the large, fibrous root system is very 

hard to pry out of the soil (Leslie M.J. Gottesfeld, unpublished notes, 1988). 



// 



Habitat considerations. —Locations where the edible fern rootstock grows in ade- 
quate abundance and quality for harvesting in British Columbia are generally 
scattered and infrequent. It is a slow-growing perennial fern and hence may be 
several years old before it is big enough to harvest. Following are some notations 
on the best habitats or localities observed for edible wood fern growth and/or 
harvesting (Native group and literature citation given in parentheses): 

near salmonberry bushes (Rubus spectabilis) (Tsimshian; Gorman 1896); 

hillsides and under cottonwood trees (Populus balsamifera) and alders (Alnus 
rubra); wooded locations up to 800 m (Tanaina; Kari 1987; Ray ca. 1980); 

half way up the mountains" in "meadows" or "ravines" (avalanche 
tracks), open snowbed communities; best locations had ownership harvesting 
claims (Gitksan, Wet'suwet'en; Jeff Harris, Sr., Beverley Anderson, personal com- 
munication to LMJG, 1988); 

plentiful around Kuldo (deserted Gitksan village about 90 km north of 
Hazelton) and around other, modern Gitksan villages; Kuldo people were teased 
in song about their consumption of the fern rootstocks (Gitksan; 'Ksan, People 
of 1980); 

at the base of big cliffs in avalanche paths; areas frequented by mountain 
goats, who survive on wood fern rootstocks during the winter (Haisla; Gordon 

Robinson, n.d.; see note 20); 

higher elevations at the base of snow banks and rockslides (Nuxalk; Turner 



1973); 

swampy areas (Mainland Comox; Randy Bouchard, personal communica- 
tion to NJT from 1973 research); 



8 



TURNER et al. 



Vol. 12, No. 1 



Fountain Valley, about 1.6 km past Rusty Creek, called sts'ets'kwa7 after 
wood fern (Lillooet; Randy Bouchard, personal communication to NJT from 1974 



researc 



nfrequent and because harvesting eliminates an entire plant 



care must be taken by those harvesting this food to leave s 
in each locality for any future harvests, or entire populat 
important food plant could be eliminated. 



Native food use of wood fern 



summary of ethnographic reports 



nomenclature for edible wood fern rootstocks is provided in Table 1 . As 
shown in this table, edible wood fern rootstocks were used by virtually all North- 
west Coast Native peoples of British Columbia, as well as by the Lower Thomp- 
son, Lillooet, Nishga, Gitksan, and Wet'suwet'en, by the southeast Alaskan and 
western Washington Indian groups, and even by some Eskimo peoples of Alaska. 
In all, folk names for this food were used in over 25 different Native languages 
in the region. Fig. 1 shows the extent of former wood fern use in northwestern 
North America. 



TABLE 1.— Summary of ethnographic reports and native nomenclature for edible 



wood fern rootstocks. 



Language 
(Family) 



Nuxalk 

(Bella Coola, 
Salish) 



Native Name 1 



Identification; Notes; Reference 



sq'Walm (rhiz.); Dryopteris expansa (most people), D. 



sq' u 'dlm-iixw >, 



iixw (plant) 



filix-mas (some people), and Athyrium 



sq'Walq'Walm- filix-femina (some people, but probably 

mistaken) (Nancy J. Turner, unpub- 



lished notes, 1981, in possession of 
NJT; Turner 1973; Harlan I. Smith, 
unpublished notes, ca. 1920-22, 
National Museum of Civilization, 

Ottawa) 



Lillooet 
(Salish, Interior) 



c '$kwa? 



Dryopteris spp. (Nancy J. Turner, 
Randy Bouchard, Dorothy Kennedy, 
Jan Van Eijk, unpublished notes, 
1974-86, in possession of NJT); 
Athyrium filix-femina said by some to 
be "a kind of c'ak^a? ," but not the 
kind with edible rootstocks (cf. coll. V 
88,796, 88,798-9, N. Turner: D. 
carthusiana) . 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



9 



Language 
(Family) 



Thompson 
(Salish, Interior) 



Native Name 1 



not recalled 



Identification; Notes; Reference 



Dryopteris spp. (Turner et al. 1990); 
eating of pit-cooked fern rootstocks 
recalled, but not identity; 
c'6kwi7, the term cognate with wood 
fern rootstocks in other languages, 
applied to fresh rhizomes of Pteridium 
aquilinum by Annie York 



Comox 
(Salish, Coast) 



th'ekwa 



Athyrium filix-femina (Randy Bouchard, 
unpublished notes on Mainland 
Comox, 1973-76); identified from 
specimens by Bill and Rose Mitchell 
(Randy Bouchard Coll. Nos. 44, 4, V) 



Sechelt 
(Salish, Coast) 



stsawch 



unidentified fern (Athyrium or Dryop- 
teris) whose rootstocks with finger-like 
appendages were pit-cooked and 
eaten (Nancy Turner and Jan Timmers, 
unpublished notes on Sechelt ethno- 
botany, 1972, in possession of NJT) 



Sechelt 
(Salish, Coast) 



xwulhqw'at 



unidentified "type of fern, something 
like a turnip," with a texture "some- 
thing like garlic, ' ' which grows in the 
mountains, especially at a place called 
xenichen (J. Joe, personal communica- 
cation to Randy Bouchard, 1978) 



Squamish 
(Salish, Coast) 



Halkomelem 
(Upriver) 

(Salish, Coast) 

Clallam 
(Salish, Coast) 



ts 'ekwa7 



th 'ikwa 



tsa 'qwa 



"Dryopteris austriaca" and, by some 
Polystichum munitum also (rootstock: 
(Bouchard and Turner 1976) 



"mountain fern with wide top 
(Galloway 1982) 



// 



'Dryopteris dilatata" (Gunther 1973) 



10 



TURNER et al 



Vol. 12, No. 1 



TABLE 1.— Summary of ethnographic reports and native nomenclature for edible 
wood fern rootstocks. (continued) 



Language 
(Family) 



Native Name 1 



Identification; Notes; Reference 



Straits 
(Northern) 

(Salish, Coast) 



tso'kwi 



(Kinkade 1989)2 



Straits 
(Lummi) 
(Salish, Coast) 



tsuk'kwa 



"small brake fern" (Kinkade 1989; 
original, George Gibbs 1863) 



Lushootseed 
(Suquamish) 
(Salish, Coast) 



ts'E'kwi 



unidentified fern (Barbara Lane, 
personal communication to NJT, 1985; 
from unpublished notes of T.T. 
Waterman ca. 1920 and J. P. Harring- 
ton ca. 1910) 



Lushootseed 
(Green River) 



tso'kwi 



"Dryopteris dilatata" (Gunther 1973) 



Lower Chehalis 
(Salish, Tsamosan) 



c 'qw{? 



Upper Chehalis c'aqw'e? 
(Salish, Tsamosan) 



"fern sp." (Kinkade 1989, original 
from J.P. Harrington field notes, 1948) 



/ / 



tiger-lily root" (Kinkade 1989) 



Sahaptin 

(Taidnapam, 

Upper Cowlitz, 

or Yakima 
dialect) 

(Sahaptin) 



ts 'kwai 



"Dryopteris dilatata" (Gunther 1973, 
who calls this language "Cowlitz;" 
M. Dale Kinkade, personal communica 
cation to NJT, 1989) 



Quileute 
(Chimakuan) 



ts 'iku>t; 

c'iqWo pat 



"fern roots" (J.V. Powell and F. 
Woodruff in Gunther 1973; Powell 
and Woodruff, Quileute Dictionary, 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



11 



Language 
(Family) 



Kwakwala 
(Wakashan) 



Native Name 1 



Identification; Notes; Reference 



from M. Dale Kinkade, personal com- 
munication to NJT, 1989) 
tsdkus (rhizome); "Dryopteris spinulosa dilatata," 
tsdganu (leaves); D. austriaca and Polystichum munitum 
tsakusmes (plant) both types of which were eaten (Boas 

1921, 1934; Curtis 1915, "wood-fern 
roots," C. F. Newcombe 3 ; Turner and 
Bell 1973) 



Haisla 
(Wakashan) 



t'ibam (rhizome); Lincoln and Rath 1986 — "edible fern 



t'ipas (plant) 



root;" "... plant, leaves" (cf. t'ipa 
"to step, tread on something"); Curtis 
1907, "wood fern" 



Heiltsuk 
(Wakashan) 



t'ibam (rhizome); Brian Compton, personal communica- 
tion to NJT, 1988 4 (cf. t'ipa "to step, 
tread onto something; to find fern 
roots or cockles by feeling with the 

feet"); Boas 1928 "tie 'bEm" -root of 

Dryopteris austriaca 



Oowekyala 
(Wakashan) 



tet-pum 



"fern-roots (wood fern)" (Curtis 1915) 



Nuu-chah-nuulh 
(Hesquiat) 

(Wakashan) 



t'ipa 



unidentified fern closely resembling 
Dryopteris austriaca and Pteridium 



same 



ago 



(Turner and Efrat 1982) 



Nuu-chah-nuulh 

(Clayoquot) 



ti-pa 



"fern-roots (wood fern)" (Curtis 1916) 



Ditidaht 
(Wakashan) 



t'it'eTsapt 



Polystichum munitum (Turner et al. 
1983); (cf. t'it'ets "crouching" for 



12 



TURNER et al. 



Vol. 12, No. 1 



TABLE 1.— Summary of ethnographic reports and native nomenclature for edible 
wood fern rootstocks. (continued) 



Language 
(Family) 



Native Name 1 



Identification; Notes; Reference 



fiddlehead shoots); rootstocks pit- 
cooked and eaten. 



Quileute 
(Chimakuan) 



t s 'ikwi 



Polystichum munitum (roots) (Gunther 
1973); Quileute said to bake rhizome 
in a pit and eat with salmon eggs. 



tseqwe' 



Athyrium filix-femina (root) (Gunther 
1973); Quileute said to eat roasted, 
peeled rhizomes 



Coast Tsimshian 
(Tsimshian) 



// 



ahh"; aa (?) 



"Aspidium spinulosum var. dilatatum 
(Gorman 1896); "an edible root; 
a root medicine' ' (Dunn 1978) 



/ f 



Nishga 
(Tsimshian) 



V 

ax 



it 



? Athyrium filix-femina" (McNeary 



1974) 



Gitksan 
(Tsimshian) 



V 

ax 



"Cystopteris bulbifera" ('Ksan, People 
of 1980); "Dryopteris filix-mas" 
(Gitanyow Summer Student Research 
Program 1984); Dryopteris expansa 
(Leslie M.J. Gottesfeld, unpublished 

notes, 1987, 1988) 



Tlingit 
(Tlingit) 



k'walx, k!wA4x 



Dryopteris expansa (Leslie M.J. Gottes- 
feld, unpublished notes 1987, 1988); 
"Dryopteris spinulosa" (Swanton 1909) 



Chilcotin 
(Athapaskan) 



?ax$ 



Nancy J. Turner (unpublished notes, 
1988); Teit (1909:780) noted that "fern- 
root" was eaten by the Chilcotin; 
Morice (1893) stated that "• ah" was 
not found in Chilcotin country 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



13 



Language 
(Family) 



Native Name 1 



Identification; Notes; Reference 



Carrier 
(Athapaskan) 



// • 



ah", 



or ah chun 



Morice (1893); second term from 
Smith (unpublished notes on Carrier, 
ca. 1920-22)-identified as "Shield 
Fern" (see note in text) 



Wet'suwet'en 
(Athapaskan) 



di yii'n 



Dryopteris expansa (Leslie MJ. Gotts 
feld, unpublished notes 1987, 1988) 



Tanaina 
(Athapaskan) 



uh 



Dryopteris dilatata (Kari 1987) 



Ha id a 

(Skidegate) 

(Haida) 



ts'agwl; djagwal; applied variously to large, edible 



skyaw ("tail") 



rootstocks of Dryopteris austriaca, Poly- 
stichum munitum, Athyrium filix-femina 
(Turner, unpublished notes on Haida, 
ca. 1971; see Note 3 in text) 



Haida 

(Masset) 

(Haida) 



t s 'agzvl; 

tanskyaw 

("black-bear's 



applied variously to large, edible 
rootstocks of Dryopteris austriaca, 
Polystichum munitum, Athyrium filix- 



tail"); sndn-djang femina (Turner, unpublished notes on 

Haida, ca. 1971; see Note 3 in text) 



Haida 

(Kaigani) 

Haida) 



sk'ydaw . 
(rhizome) 



"Thelypteris limbosperma" (Norton 
1981); specimen later reidentified by 
Adolf Ceska as Athyrium filix-femina 



Western Eskimo wingisuk 
(Inuit) 



1 



Dryopteris dilatata (Oswalt 1957) 



Orthography used in original source is given, except j$ 



?;£ = x . 



x. For simplicity, 



not all reported terms for Athyrium filix-femina or Polystichum munitum are included in this 
table, only those cognate with terms applied to Dryopteris spp. or directly associated with 



reports of edible rootstocks. 



2 



nication to NJT, 1989) notes that the name for spiny 



M Dale Kinkade (personal commu . 

wood fern is reconstructable in Proto-Salish, and hence provides evidence for coastal origins 

for Salish (cf. Kinkade 1989). 



14 



TURNER et al. Vol. 12, No. 1 



^C.F. Newcombe gives several variations of the term tsakos in various parts of his unpub- 
lished notes and manuscripts (Newcombe Coll. Vol. 43, File 36, ca. 1922, Provincial Archives 
of British Columbia); at least two were confirmed by Kwakwala speaker and linguist George 
Hunt; all refer to "Dryopteris Ailatata" or its synonyms. 

4 Brian Compton compiled a comprehensive list of fern names for Kwakwala and related 
languages, entitled "North Wakashan Pteriodophyte Nomenclature & Terminology" (1988), 
which contains a much more complete list of fern species and related terminology in Kwakwala, 
Haisla, Heiltsuk, and Oowekyala (Rivers Inlet) (ms. in possession of Brian Compton, Depart- 
ment of Botany, University of British Columbia, Vancouver, BC). 

^This name was given by Robert Tyhurst (1975-76, unpublished notes on Chilcotin ethno- 

botany, in possession of RT, Victoria, British Columbia) as bracken fern (Pteridium aquilinum), 
but recent evidence suggests a Dryopteris species, from the descriptions of the clustered 
root stocks. 



Many descriptions exist of the edible 



clasped together 7 ' (Norton 1981; B. Lane, personal communication 
a woody sweet potato' ' ('Ksan, People of 1980), "resembling a bunch 



like 



// 



like a pineapple" (Harlan I. Smith 



Matthews, Haida elder, tape transcript by Nancy J. Turner, ca. 1971 
many brown "fingers" growing around it (Turner 1973). The "good 



(Will 



succulent. Figures 2-4 show the characteristics of the edible wood fern 



Dryovt 



Harris Sr. described the edible 



as big as his hand (fist), and noted," . . when you take them out the root tapers 
down to the bottom, when you dig it out, and crooked. You have to take the 
little piece of the bottom part out because it's small . . . that [leaf base] would 
be about the size of your little finger ... the banana-like root. You take them 
off and peel it with your finger ..." (Leslie M.J. Gottesfeld, unpublished notes, 
1987). The fern rootstock is described as being greenish inside when raw, but 
turning to yellowish or orange when cooked (Jeff Harris, David Green, Lizette 
Naziel, personal communications to LMJG, 1987). 

Harvesting. -The rootstocks were usually dug in spring or fall (cf . 'Ksan, People 
, The Nuxalk name for the fourth ^^^ ^ summer so i st i C e is called 



siq aalxmantn (lit. "time 



time, many of the "fingers 



// 



rootstock are plump and round, whereas earlier many are flat and no good 
ior eating. Potatoes are dug at this time too, according to Margaret Siwallace 
(Turner 1973). However, the wood fern rootstocks could be dug out even in 
December if it was a mild winter, and could be dug from under the snow if 
necessary. Wet'suwet'en consultants stated that wood fern rootstocks should be 
gathered in fall after the leaves wither or in the winter (Josephine Michell 1987, 
Madeline Alfred 1988, Lizette Naziel 1987, Sara Tait 1988, all personal communi- 
cations to LMJG). It was stated that the rootstocks are not damaged by freezing 
and could be dug from under the snow. A special wooden snow shovel was 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



15 




FIG. l.-Extent of former use of edible wood fern (mainly Dryopteris expansa) 
in northwestern North America. (The heavy dotted line shows the approximate 
distribution of Dryopteris expansa over the same area.) 



16 



TURNER et al. 



Vol. 12, No. 1 







FIG. 2. 



from previous 



Dryopteris expansa, showing ' 'finger' ' formation of leaf 



// 



whitish 



used to uncover the plants (Madeline Alfred, personal communication to LMJG 
1988) . Gitksan consultants stated that fern rootstocks were gathered in fa 
or spring before the fronds begin to uncurl . Plants were located by the 
stick sticking up" or the "curly leaf" (fiddlehead) . Digging the rootstocks out 
of frozen ground was laborious, but their availability in the winter season made 
them a valuable food. Some Tanaina people say they can be dug at any time of 
the year, but others say that they are dry and unpalatable during the summer 
months. Some say the rootstocks are juiciest and most palatable in the fall, others 
in the spring (Kari 1987). 

Since the rootstocks were usually sought after the tops had died down, 
finding them was sometimes difficult. The Haisla and Heiltsuk people evidently 
had a unique way of locating them with their feet, as alluded to in their term, 
ftpm, meaning "to step, tread onto something; to find fern roots or cockles by 
feeling with the feet" (Heiltsuk), and in ftps, meaning "one's feet touching the 
ground (as when feeling for fern roots)" (Heiltsuk) (Brian Compton, personal 
communication to NJT, 1988; terms given to him by Kitlope elder, Gordon Robert- 
son; see also Lincoln and Rath 1986 for Haisla). It is not clear whether the actual 
names for the edible wood fern are derived from these terms, or vice versa. 21 
According to Gordon Robertson (Brian Compton, personal communication, 1988), 
a special digging stick was used to dig up wood fern rootstocks. It was similar 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



17 




FIG. 3.-D™ 



FIG. 4. 



tpansa, rootstock and fronds, taken at Bella < 
Nuxalk elders as the edible type of wood fern 
m of Dryopteris expansa rootstock (right). 



18 



TURNER et al. Vol. 12, No. 1 



to the stick used for other edible "roots," but shorter (about 30-45 cm 1 
with a flattened point and a round portion on the upper end similar 
head. The stick, called cagayu in Haisla, was pushed into the groi 

nd then pressed down like a lever to remove the root from 



Recently, 



purpose 



use of a digging-stick, of yew-wood, for prying up the fern rootstocks, wl 
were then placed into a large basket. The Tanaina generally obtain wood : 
rootstocks by chopping them out of the ground with an axe (Kari 1987). 

Harvesting and cooking of the wood fern rootstocks, like the harvestin 
most plant foods, was apparently undertaken mainly by women (cf . Boas 19 
but men also dug them ('Ksan, People of 1980). 



Storage. —Harlan Smith (unpublished notes, ca. 1920) maintained that wood 
rootstocks "... were always cooked [by the Carrier and Nuxalk (Bella Co 
the same day they were gathered. They were never kept." Morice (1893) n 
that the rootstocks were not dried, but only eaten fresh after pit-cooking. ( 



Carrier consultant. 

The Wet'suwet 
harvesting them if r 
stated: 



mashed, the food only 



cording to Smith 



? women] would get together and leave Hagw 
. Above Blue Lake . . . they would dig a whole bi 
Then thev would build fires. . . . Thev would 



put 



the fruit [rootstocks] under the rocks to cook it. They would pres 
it for the winter. They lined cedar boxes with skunk-cabbage leaves 
[lysichitum americanum] and preserved all that fruit in there. They sealed 
the boxes up and dug the ground like cellars. . . . They would preserve 
all those things for winter, put them away and save them until the hungry 



time 



communication to LMJG 



The Haisla of Kitamaat Village also formerly stored fern rootstocks for winter 
Gordon Robinson recounted: 



In the old days each family would pick ten sacks, or baskets of t'ibam 
[fern rootstock]. They would bury it in the corner of the long house. Like 
a root cellar. They break the base off. It is a useless part. They pick it 
in the fall . . . (personal communication to LMJG, 1988). 

The Kwakwaka'wakw (Southern Kwakiutl) spread wood fern rootstocks out 
to dry on a mat the day after they were harvested, then cleaned them using a 
special red-cedar-wood scraper, and stored them in a basket in the rear of the 
house, behind the fire, for 12 days before they were pit-cooked. After they 
were cooked, they were stored another four days before being served at a feast 
(Boas 1921). The Mainland Comox also stored them prior to cooking in a cool, 
dry place, in an open-work basket (Randy Buchard. unoublished notes. 1973) 22 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 19 



The Gitksan, too, commonly store edible fern rootstocks. After they were dug, 
they might be left in a dry shed or cabin, covered with brush and leaves, for a 
long period of time, "when other food supplies ran out" ('Ksan, People of 

1980:79). 

In Alaska, Tanaina people are said to preserve wood fern rootstocks by plac- 
ing them in an underground cache or by storing them in oil or lard (Kari 1987). 
The Yakutat people also stored them in a pit cache, and in general stored them 
in a way similar to potatoes. They also keep well for several months in a 
refrigerator, according to Ray (ca. 1980). 



cooking and serving methods. —Wood 



Nuxalk people sometimes ate them raw as an antidote 
shellfish poisoning.23 Baking or steaming the rootstocks in a pit was by fai 
most common traditional method of preparation. Detailed published account; 
it-cookine them are provided in Gorman (1896:78-79) and Boas (1921:518-523) 



In the former description, the Coast Tsimshian 



moss or kelp and cooked them 



overnight (about 14 or 15 hours). In Boas's Kwakwaka'wakw (Southern Kwakiutl) 
account, seaweed and hemlock branches (Tsuga heterophylla) were used to sur- 
round the rootstocks, which were also cooked overnight. Members of the 
household were asked abstain from sexual intercourse during the cooking time. 
Similar descriptions of pit-cooking wood fern rootstocks are provided by many 
others (cf. Morice 1893; Boas 1921; Harlan I. Smith, unpublished notes on Carrier 
and Bella Coola, ca. 1920-22; Nancy J. Turner, unpublished notes on Haida, 1970, 
1971; 'Ksan, People of 1980; Leslie M J. Gottesfeld, unpublished notes on Gitksan, 
Wet'suwet'en, and Haisla, 1987, 1988; Jacobs and Jacobs 1982; Kari 1987). Gitksan 
elder Jeff Harris Sr. said that the rootstocks were arranged in the pit ". . . Just 
like putting apples" [placed upright, in growth position] and were covered with 
hemlock boughs (personal communication to LMJG, 1987). The Wet'suwet'en 
covered them with birch bark (Leslie M.J. Gottesfeld, unpublished notes 1987). 
The Tanaina sometimes wrapped them in birch bark for baking and covered them 
with hot sand (Kari 1987). Haisla people used to cover the rootstocks with hot 
embers of a fire and leave them to bake overnight (Gordon Robinson, personal 
communication to LMJG, 1988). The Haida used to line their steaming pits with 
skunk-cabbage leaves. The Carrier were said to use alder bark chips in the steam- 
ing pit (Morice 1893). The Mainland Comox cooked the rootstocks they call 
th'ekwa (identified as Athyrium filix-femina; R. Bouchard Coll. No 44 and No. 
004; May 1975, May 1976 V) in a pit, but simply by throwing them directly into 
a bed of glowing hot coals and ashes from a fire and covering them with several 
inches of ashes (Randy Bouchard, unpublished notes on Mainland Comox 
ethnobotany, personal communication 1973, 1975 to NJT). In recent times, people 
have boiled the rootstocks in water for a long time, until they are tender; 
sometimes the water is changed during boiling (Kari 1987). One Hydaburg woman 
suggested pressure cooking them to reduce the cooking time (Ray ca. 198U). in 
Bella Coola today, pressure cooking is the method of choice for preparing wood 
fprn r^fcf^vc wL.ctc TV,p „c P of the rhizomes as a soup thickener was sug- 



20 



TURNER et al. Vol. 12, No. 1 



gested by Ray (ca. 1980). Some Gitksan people now preserve the rhizomes by 
canning ('Ksan, People of 1980). When not properly cooked, the rootstocks are 
described as rubbery and hard (Mark Jacobs Jr. of Sitka, personal communica- 
tion to LMJG, 1988) . 

Boas (1921:523-524) provides a detailed description of the serving of wood 
fern rootstocks at a Kwakwaka'wakw (Southern Kwakiutl) feast, which was held 
four days after the fern roots had been in the house and had been cooked. The 
roots were considered a "really valuable food," and were often served with oil 
and dry silver-salmon spawn to the chiefs of the tribes. A chief could peel and 
eat the outer fern roots, but was supposed to give away the inner part of the 
fern root, not to eat it himself, or "he will always waver in his mind about giving 
away blankets ..." After the feast, each guest was given two fern roots to take 
home to his wife. 

There are many other reports of the rootstocks being cleaned, the leaf bases 
being removed, peeled, and eaten one at a time. Similarly, many reports refer 
to the eating of the rootstocks with oil, grease, lard, and/ or salmon roe (cf. Nancy 
J. Turner, unpublished notes on Haida, 1971; Turner 1973; Randy Bouchard, 
personal communication to NJT, 1973 for Mainland Comox; 'Ksan, People of 1980; 
Kari 1987; Brian Compton, personal communication to NJT, 1988, for Haisla; Leslie 
M.J. Gottesfeld, unpublished notes on Gitksan, Haisla 1987, 1988). The Haisla, 
for example, ate them mixed with ooligan grease in a big bowl (Brian Compton, 
personal communication to NJT, 1988, from Gordon Robertson), or ate them with 
fermented or dried salmon roe (Gordon Robinson, personal communication to 
LMJG, 1988). The leaf bases, once removed from the main rootstock, were usually 
peeled with the fingers before being eaten, then dipped into the oil. The Mainland 
Comox dipped them into seal oil and ate them with dried salmon eggs (Randy 
Bouchard, unpublished notes on Mainland Comox ethnobotany, personal com- 
munication to NJT 1973, 1975). The Nuxalk ate them with grease or fermented 
salmon roe (McDwraith 1948; Turner 1973). In modern times, the Gitksan ate them 
with sugar ('Ksan, People of 1980). 

Kari (1987) notes that some Tanaina people state that onlv the "stem" por- 



both 



the 



say both "stem 



pit-cooking the rootstocks, pounded them up 



stones, the upper one apparently a pestle, and the lower one flat (Harlan I. Smith, 
unpublished notes on Carrier and Bella Coola, ca. 1920-22). Smith noted that 
"Sometimes the Ulkatcho Carrier Indians went to Bella Coola and traded moc- 
casins for pestles and other kinds of goods 



Oswalt (1957) reported that the rootstocks of "Dryopteris austriaca" were 



cream. "24 



water 



// 



Eskimo 



Sometimes the rootstocks were simply roasted in the fire, covered with 
le, if there were not enough to pit-cook ('Ksan, People of 1980). 



Other food uses. —The Tanaina and other southeast Alaskan natives eat the young 
croziers, or "ftddleheads" of wood fern (uh), cooked or steamed, as an early spring 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 21 



vegetable, and many southeast Alaskans can them for winter use (Kari 1987; Heller 
1976). The Kaigani Haida are said to boil and eat the "fiddleheads" (of 
"Thelypteris"; see Table 1) as a vegetable at present (Norton 1981). According to 
Ray (ca. 1980), fern fiddleheads, commonly eaten among many Native peoples 
of Alaska today, are not known to be a traditional food. He provides recipes and 
suggested methods for canning and freezing these greens. Some Tanaina people 
also used wood fern rootstocks for making beer, a practice they probably learned 
from the Russians, according to Kari (1987). 25 



Survival and famine food. —As recounted by Peter Kalifornsky of the Outer Inlet 
Tanaina (Kari 1987:130), the rootstocks were formerly regarded as a good sur- 
vival or starvation food: 

In the early spring one year, the people ran out of food. They 



groups, one moving 



name 



clams 



weieht. while those that lived on clams 



In winter and early spring, the rootstocks had to be chopped out of the ground, 
after a fire was first built over an area where the ferns were known to grow in 
order to thaw the ground (Kari 1987). The Tlingit (Jacobs and Jacobs 1982) and 
Gitksan ('Ksan, People of 1980) also state that wood fern can be a survival food: 
"It has warded off starvation more than once." ('Ksan, People of 1980:79). Mark 



Jacobs Jr. of Sitka (personal communication to LMJG, 1988) stated that perhaps 
fern rootstocks were eaten sometimes as an ordinary food, but they were mainl) 
used as a survival food by the Tlingit. Gitksan and Wet'suwet'en stories also refei 
to fern rootstocks as a survival food. A woman who was banished fron 
Moricetown survived through the winter on fern rootstocks and salmon ro< 
according to a Wet'suwet'en story (Alfred Joseph, personal communication tc 
LMJG, 1986). Jeff Harris Sr. mentioned a Gitksan legend about a period of star 
vation: "There in the famine some people eat ax [fern rootstock]. They lived 
Others go to get xsu'u [hemlock cambium] but they all died." (personal com 



munication to LMJG 



frequency and taste appreciation of wood fern rootstocks. -An interview study 
1 adult Nuxalk women in three generations, representing the Nuxalk reserve 



in 



Programme. Each woman was asked for her family's use for wood fern root 
(and other traditional food species as well), and a score of her impression of her 
own personal taste appreciation of the cooked root "fingers." Of the 61 women, 
only six (all elders) reported ever tasting the root food. On a scale of 1.0 (not 
used) to 4.0 (used more than one time per week when available and in season), 
the women reported a mean use frequency today of only 1.3 (very low) but in 
their earlier days they used it to a greater extent, with a score of 2.8 on a scale 
of 1.0 to 4.0, which represents use of approximately once or twice a month. These 
women reported never to have preserved or stored the roots (in cellars or other- 
wise) for later use. 



22 



TURNER et al. Vol. 12, No. 1 



The mean taste appreciation score of the six women was 4.3, in a scoring scale 
of 1 to 5, with 5.0 being the highest possible (best tasting food), indicating these 
women enjoyed this food. Details of the methodology are reported elsewhere 
(Kuhnlein 1989a). 

Gorman (1896:79) described the taste of wood fern rootstocks as, "slightly 
sweetish'' but "too smoky and tobacco-like in flavor for the average white man's 
palate, except under stress of hunger." He added, "... I have no doubt it is 
quite nutritious." The Tanaina also considered uh to be very nourishing, as 
indicated from their previously related use as a survival food (Kari 1987). Jacobs 
and Jacobs (1982) describe the edible portion as being "light brown like squash 
and tastes about the same." Haida elder Willie Matthews described the cooked 
"fingers" (leafstalk bases) as "just like potatoes inside" (Nancy J. Turner, unpub- 
lished notes, 1971). Ray (ca. 1980) noted that the raw rhizomes of the wood fern 
("Dryopteris dilatata") are bitter, but sweet tasting when cooked. He said they 
are salmon colored, with texture and taste very similar to sweet potato. The 
Gitksan people also describe the edible fern as having the same texture and color 
as a sweet potato when cooked, and believe it to have "considerable food value" 
('Ksan, People of 1980:79). One Lillooet elder described the taste as "very much 
like coconut" (Randy Bouchard, personal communication to NJT, 1974). 



Nutrient 



summer and early fail of 



under the 



Food and Nutrition Programme, approximately 30 wood fern n 
with the supervision of Nuxalk elder Felicity Walkus. These were cleaned in the 
traditional manner by washing with water and then (for convenience) cooked 
by pressure cooker until soft. The cooked roots were then peeled, and 250 g were 
frozen together in a plastic bag and shipped to the laboratory. A similar sample 
was prepared in 1983. 

The laboratory samples were mixed with equal weights of distilled, deio- 
nized water, blended in a glass container with stainless steel blades and processed 

arious nutrient analyses. The details of these methods are reported in 



Kuhnlein 



nutrient analyses of the wood fern "root" samples are prese 



in table 2, together with reported values for the common potato, baked in the 
skin. It can be seen that there is reasonably good agreement between the two 
foods for water, protein, ash, carbohydrates (computed by difference between 
total weight minus the sum of protein, fat, moisture, fiber, and ash) and approx- 
imate energy computations. In contrast, mineral values in fern "root" were much 
higher for calcium, magnesium, zinc, copper, and manganese, but higher in potato 
for sodium and iron. The differing values may have been in part due to contri- 
butions from the skin of the roots, since it is known that skins of root foods are 
more mineral-rich than the starchy flesh. These minerals can migrate into the 
flesh during cooking. 

The proximate composition of fern "root" is in good agreement with that 
reported for Trifolium wormskioldii (springbank clover rhizomes) 



com 



; fica (Pacific silverweed roots) (Kuhnlein et al. 1982). For mineral 
fern "root" was slightly higher in calcium, phosphorus, and 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 23 








TABLE 2.-Nutrien 


ts in 


cooked 


fern root (ur 


yopteris 


expansa) in 


comparison to 




cooked potato, per 


100 


g edible 


portion. 










Nutrient 








Fern Root* 


Potato** 




Water, g 










68.4 


71.2 




Protein, g 
Fat, g 

N.D. Fibre, g 
Ash, g 

Carbohydrate, g 
Calcium, mg 
Phosphorus, mg 
Sodium, mg 
Magnesium, mg 
Iron, mg 
Zinc, mg 
Copper, mg 
Manganese, mg 










2.5 
1.0 
3.7 

0.8 
23.6 
56.3 
62.6 

1.4 
44.4 

0.8 
1.5 
1.5 
3.2 


2.3 

0.1 

n.a. 

1.2 

25.2 

10.0 

57.0 

8.0 

27.0 

1.4 

0.3 

0.3 

0.2 




*n = 2 independent 


same 


ties, analy 


zed in triplicati 


e. 







**Watt and Merrill (1975:106), potatoes baked, flesh and skin, n = 12. 



magnesium. However, fern root contained zinc, manganese 



m 



silverweed roots, clover rhizomes, or potato (ranging from 
0.8 mg for these three minerals in the three foods). 

While nutritionally essential minerals are important to consider in the: 
foods, it must be kept in mind that the majority of minerals in the trad 
diet are provided on a year round basis in fish and game. In contrast tc 
animal foods, the roots provide a minimal proportion of daily mineral i 



ethnographic accounts give the impression 



were 



harvesting 



peeling of the small "fingers" before eating. With 



mind, one has to assume 



major proportion of the bulk, or of the energy value of 



could be relied on as a famine 



food, since the ferns were accessible even in winter to those who knew where 



them 



Medicinal uses of wood fern and related types. -A significant use by the Nuxa k 
of Dryopteris expansa ("Aspidium cyclosomm") was eating the rootstock raw to 
neutralize poisoning from eating several kinds of shell-ftsh in the early part 



24 



TURNER 



Vol. 12, No. 1 



of the summer (presumably paralytic shellfish poisoning, but not proven as such) 
(Smith 1928). The physiological basis for this application is not clear. Possibly, 
the information relates to the use of Dryopteris filix-mas rootstocks as an anthel- 
mintic or vermifuge drug (Claus et al. 1970). Incidentally, we could find only one 
reference to use of a fern as a vermifuge by Native peoples in the study area: 
Smith (1928) noted that the Northern Carrier boiled the root of a fern ("species 
uncertain 7 ') and drank the decoction for worms. 

Most other medicinal uses of wood fern and its relatives were as poultices 
for skin ailments. The Clallam of Washington pounded the roots of ''Dryopteris 
austriaca" and used the pulp as a poultice for cuts (Gunther 1973), and the 
Inland Tanaina boil the rootstocks of uh (apparently including both Dryopteris 
expansa and Athyrium filix-femina) in water and use the liquid as an eye wash and 
as a wash for cuts. The decoction is also drunk for tuberculosis, kidney troubles, 
and respiratory problems such as asthma (Kari 1987). The Haida used the boiled, 
mashed rootstocks of sword fern (Polystichum munitum), which were called by 
the same Haida name as "Dryopteris austriaca" by several Native consultants, as 
a poultice for cuts and swellings (Nancy J. Turner, unpublished notes, 1971). 



Other, miscellaneous uses.— The thin, wiry roots of "Dryopteris austriaca" were 
sometimes used by the Kwakwaka'wakw (Southern Kwakiutl) as the burning 
material in a "slow match." They were enclosed within a clam or mussel shell 
and ignited. The shell could then be buried and the fern roots would smoulder 
for several days (Turner and Bell 1973). 

The Upriver Halkomelem people sometimes gather the fronds of the "moun- 
tain fern" (Dryopteris sp.) for use by florists (Galloway 1982). The Western Eskimo 
of Alaska have used the fronds of ' 'Dryopteris austriaca ' ' in recent years to decorate 
the inside of the church and the graves in the cemetery on certain church holidays. 
Additionally, ferns were described as having adorned some of the ceremonial 
masks used by the Kodiak Island Eskimos (Oswalt 1957) . 

Gunther (1973) notes that the Snohomish (Lushootseed) soaked the leaves 
of Dryopteris austriaca for a hair wash. The Haida used the dried fronds for bed- 



unpublished notes, 1971). 



drying fish to prevent molding (Nancy J 



Mythology and traditional beliefs. —Edible wood fern is featured in several myths 
within the areas where it was used. There are at least two myths explaining the 
origin of this food, one Tlingit and one Lower Lillooet. In the Tlingit account, 
the head of an orphan girl entrapped by a landslide near her village was trans- 
formed into a fern root (k.'wAh), while her body became a ground hog (marmot) 
(Swanton 1909:180).26 l n the Lillooet episode, part of a much longer, more 
complex story called Kaiydm , a blind old woman was transformed into a wood 
fern plant by her angry husband because she had allowed their grandson to be 
stolen. He took her up and jammed her nose into a log, saying . . ., "Then 

shall become a Tsukwa [wood fern]. By and by people will eat you ..." 
Tout 1905:177).27 



(Hill 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 25 



There are two other short episodes in Tlingit mythology concerning the wood 
fern. One is about its transformation by Raven, told within a long story of how 
Raven changed people, animals, and plants into their present form (Swanton 
1909:18). The other is about a man who was aided in becoming a hunter by a 
supernatural mountain being. The man was warned that the green fern roots, 
which grow wherever there are grizzly bears, were not to be used because they 
belonged to the mountain being (Swanton 1909:358). Other mythical accounts 
of wood fern have not been previously published. 

The following story about Sq' w alm, Wolverine, and Raven is shared in 
Kimsquit (Nuxalk, or Bella Coola) and Heiltsuk mythology, according to the late 
Margaret Siwallace, whose father, Joe Saunders, had told it to her around 1940: 



Wolverine went up to a certain mountain [above Kimsquit] 



mountain 



moun 



tain goat and people (he is half goat , half person, sort of like the my 
centaur) came along. "What are you doing?" he asked. Wolverint 
"Roasting mountain goat kidney." Then the goat-person tested his 
He shot in several directions and each time he shot a mountain go. 
peeled away the fat from around the stomach of each and gave 
Wolverine. Wolverine took it home and gave it to his children, wh 



time 



wanted some, but thev were not invited to eat. Wol 



told Raven how he got it. 

The next dav. Raven decided to get some mountain 



some 



came. However 



Raven answered the truth— roasting sq'Walm. Goat-person did not test 



mountain eoat fat from him 



mountain 



own breast open and removed the fat from 
children telline them it was mountain goat fat 



When 



(The late Margaret Siwallace, personal communa 



NJT 



story 



) Gitksan elder Jeff Harris Sr. people used to tell stories about 
community house. He has given permission for the following 



There was a family of four: a man and his wife and two children. 
The older one was a girl and the small one a boy of three or four. When 
they ran out of food in the spring time there was still snow upon the 
"ravines." They used a wood paddle or shovel or spade, sin t'ul, to scrape 
aside the snow [on the "ravine"]. They find ax, The stems still come up 
under the snow. They gather it all up and roast it under fire. 



26 



TURNER 



Vol. 12, No. 1 



The girl is looking after the little boy at the camp while the mother 
is [about 300 m] from the camp. The little boy cries. The girl is looking 
after the boy. The girl hollered to her mother, "The baby is crying." The 
mother replied, "Mi 'ooda his todzin. Mi' 'ooda." She meant, put some 
ax in the ashes, prepare a bit of food for the boy. 

The girl got it wrong and threw her brother in the fire. She hollers, 
"I threw him in the fire but he kept on coming back!" The boy got burned. 

The story continues with an account of how a stranger dressed in black, a t 
person, appeared and told the family how to use bear oil or fat on burns as 
ointment or medicine. Harris explained that rendered bear fat could be usee 
grease when eating fern root, which may explain the linking of the story of 
bear offering his fat to the eating of fern root. 

The importance of wood fern rootstock as a winter food for the Gitksa 
indicated by its adoption as a major crest by the house of Woxsimlaxhaa of 
Gisqaast (Fireweed Phratry) from Kispiox. The crest is called Wit Ax ("giant w 
fern rootstock"). It refers to a story about a giant wood fern root which 1 



Kisgegas by a man after being mistaken for 



mornin 



the giant ax was discovered in the ashes. (This may be a reference to the discovery 
of how to cook Dryopteris rootstocks.) The present holder of the crest is Alvin 
Weget of Kispiox village, who inherited the crest and a button blanket which 
displays the crest (Alvin Weget, personal communication to LMJG, 1988). 

Photographs taken by Marius Barbeau (1929:86-87) show two totem poles from 
Kispiox village in the 1920s bearing the "mountain fern crest" (Wii Ax). The first 
(Fig. 5) shows a geometric pattern of diamonds purported to be the "mountain 
fern crest" (wii ax). ) probablv the spinv wood fern. Modern residents of the village 



The 



Wii Ax crest in more 



representational fashion: a series of upward pointing fingers in several tiers 
topped with four or five stylized fiddleheads (Fig. 6). 

Boas (1932:227) reports a relevant traditional belief held by the Kwakiutl: 
"When fern roots (Dryopteris spinulosa, tsa'k'us) are steamed in an oven, an 
immature girl, if possible the daughter of the woman who cooks the roots, must 
tramp on the mat to press the whole down firmly." In a previous description 
of pit-cooking the ferns (Boas 1921:521) he stated that the filled cooking-pit should 
be trampled by "a woman who has had just one husband, and whose husband 
is still alive, and who has never been a widow, and whose monthly period 
terminated at least eight days before." He also noted that all members of the 
household in which the fern rootstocks were being pit-cooked should abstain 
from sexual intercourse during the night. 



CONCLUSIONS 



Although several different botanical identifications have been published for 
the edible wood fern rootstock, used as a food and emergency ration by Native 
peoples in western North America from Washington to Alaska, evidence sug- 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



27 




FIG 



from Kispiox ("Kispayaks") Village, showing 



pattern of diamonds purported to be the "mountain fern crest" (Wiiax) 
Dryopteris expansa rootstock (left). Modern 



to recognize this crest. FIG. 6. 



from Kispiox Village, showing the Wit 



more representational fashion: a series of upward pointing nngers in 
:iers topped with four or five stylized fiddleheads (right). 

Figure 5 photo by CM. Barbeau, taken in 1920. 

Courtesy of National Museums of Canada, Canadian Museum of Civilization Ottawa Ontario; Neg_ No. 4928< 

Figure 6 photo by Wilson Duff, taken in 1952. 
_ . , .. n ■ o_:.:.i. /-„i w„ m, ■<=».. m Virion* Rrilish Columbia on No. 7037 



28 



TURNER et al. Vol. 12, No. 1 



gests that the major species involved was Dryopteris expansa, a member of the 
Dryopteris austriaca complex. Within this species, the largest specimens were pre- 
ferred and selected. Other species with similar appearing rootstocks, including 
D. carthusiana and D. filix-tnas, and Polystichum munitutn, were also apparently 
eaten on occasion. Although Athyrium filix-femina has often been suggested as 
the edible wood fern, most knowledgeable Native elders who have examined its 
rootstock closely reject it as an edible type. 

Present use of wood fern rootstocks as food is extremely low. In most cases, 
only Native elders in their sixties or older remember having eaten this food, 
or have themselves gathered and prepared it. A few elders are interested in 
obtaining and cooking it "for old time's sake," and to enjoy the remembered 
flavor. Everyone universally comments on the greater convenience and availability 
of substitute carbohydrate foods like potatoes and turnips. Some potential use 
for a survival food or an infrequently used cultural specialty food seems appro- 
priate. 

Because the distribution of the edible wood fern is scattered, and because 
harvesting the rootstock for food means destruction of the entire plant, conser- 
vation of wood fern populations should be a primary consideration in any 
program involving future food use of wood fern. 



NOTES 



1 



J.P. Harrington's notes (ca. 1910) are at the Smithsonian Institution, Washington, D.C. 
(Reel 15, frame 0472). Ethnographer T.T. Waterman confirmed the same term for this 
fern about a decade later (Barbara Lane, personal communication to NJT, 1985). 

Four different unpublished manuscripts of Harlan I. Smith are cited. They contain 
unnumbered pages, many of which are duplicated under one or more different titles. 
Handwritten notes have been added to typed text with occasional dated entries. Com- 
ments on Pteridophyte data date mostly from 1920-1922. The original manuscripts 
are all held by the National Museum of Civilization in Ottawa, Ontario (formerly 
National Museum of Man), and are entitled: "The Material Culture of the Carrier 



Starvation:" "The 



Carrier Indians of British Columbia. Part 



The 
The 



and 



Joshua Moody (Nuxalk 



(Nuxalk 



Ulkatcho Carrier speaker Charlie West). Voucher specimens made by Smith 



Herbarium 



They 
3 The following unpublished notes on the Haida and Nuxalk made by Nancy J. Turner 



and in her possession, are cited in the manuscript: Turner 1970 (Haida), Turner 1971 



// 



(Haida), Turner 1981 (Nuxalk), Turner 1983 (Nuxalk), Turner 1988 (Nuxalk). 

Harlan I. Smith's unpublished notes, ca. 1921: "Lady Fern, Asplenium cyclosorum 
(collections made by Smith: 8a June 7, 1920 [CAN 525 607]; 8b June 22, 1920 [CAN 
525 608]); both were confirmed by Adolf Ceska as Athyrium filix-femina). 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 29 



^Harlan I. Smith's unpublished notes: "Shield Fern, Aspidium spinulosum" (collections 
made by Smith 7(54M) June 18, 1921 [CAN 525 615], and 7(77M) July 11, 1921 [CAN 
525 616]. Both identified by Adolf Ceska as Dryopteris expansa.) 

6 Harlan I. Smith's unpublished notes: " Aspidium filix-mas" (Smith Coll. 7(55M), June 18, 
1921 [CAN 539 343]; identified by Adolf Ceska as Athyrium filix-femina, but specimen 
is the top part of an old frond, easily mistaken for Dryopteris filix-mas— AC). 

7 This term, salidana, is sometimes applied as a general term for ferns in North Wakashan 
languages such as Kwakwala (Southern Kwakiutl) and Haisla, or to Polystichum munitum 
specifically (cf. Turner and Bell 1973; Lincoln and Rath 1986; Brian Compton, personal 
communication to NJT, 1989). 

Two ferns, one called da sun a chiin and one, ah, au, ah chun, or all chun, are men- 
tioned in Harlan I. Smith's unpublished Carrier notes. The former term, listed under 
"Lady Fern (Asplenium cyclosorum Rupr.)," has a hand-written notation: "Leaves 
coarse and less bifurcated than the shield fern." The latter, under "Shield Fern," 
states: "... has spores and is finer and more bifurcated than \ Asplenium] . . . This 
Shield fern roots used for food found 20 miles away [in the Bella Coola Valley] . . . 
roots called ah. ... it always has little [hairs?; can't read handwriting] like where 
branchlets come off stem . . . there are none on [da sun a chun}." Smith noted that 
ah chun was the same as squalum in Bella Coola, but also reported that one of his 
Native consultants (H.I.S.) could not distinguish between the two types. 

y Another aspect of the confusion is seen in Franz Boas' s classic work, The Ethnology of 
the Kwakiutl (Boas 1921), where meticulously reported accounts are given in English 
and Kwakwala of the harvesting, preparation, and serving of Kwakiutl foods. Cultural 
features of at least three edible fern species are described, all called in the English 
version simply "fern-root:" bracken (Pteridium aquilinum) —saguma; licorice fern (Poly- 
podium glycyrrhiza)-4Ek!wa e ye; and wood fern (Dryopteris expansa, or "D. spinulosa") 
tsak m use. One must consult the Kwakwala text to learn which species is being dis- 
cussed. In fact, in at least one case (Boas 1921: 526), the scientific identification for the 
fern being discussed, given as Dryopteris spinulosa, does not match the Kwakwala name 
used,-HEk!wa e ye, and the information given also matches Polypodium, since it refers to 
the "fern-root" being held in the mouth of a hunter to alleviate hunger and thirst, 
a common practice for this latter species. 

10 A small, related species of rocky areas, P. imbricans (D.C. Eaton) D.H. Wagner, some- 
times included as a form of P. munitum, extends in our area only to the eastern part 
of Vancouver Island and the Sechelt Peninsula; its rootstock is very small, and its 
use as food is unlikely. 

11 Another fern, whose fiddleheads were eaten by Native peoples of the Maritimes, 
Matteucia struthiopteris (L.) Tod. (ostrich fern), also occurs in some locations where 
edible wood fern rootstocks are used (i.e., Kispiox River and north of Kitimat) but its 
range is quite restricted, nor, to our knowledge, has it ever been cited in the eth no- 
botanical or ethnographic literature of western North America as having been eaten here. 

^Dryopteris fragrans (L.) Schott. (fragrant shield fern or fragrant cliff fern) can be excluded 
from the species possibly used as food, because of its small size and its limited distribution 
on dry limestone cliffs in northern British Columbia and Alaska (cf. Scoggan 1978; 
Hulten 1968). Dryopteris arguta (Kaulf.) Watt (coastal shield fern) is likewise excluded 



30 



TURNER et al. Vol. 12, No. 1 



because it is restricted to southeastern Vancouver Island (near Nanoose Bay) and the 
adjacent Gulf Islands (Straley et al. 1985), whereas the major reports of use of edible 
wood fern are outside of this range. Dryopteris cristata (L.) A. Gray is a relatively rare 
species of the British Columbia interior, and has not been collected at all along the 
coast (Straley et al. 1985). 

l^Some authors, including Walker (1961), questioned correctness of the name D. carthu- 
siana, but we followed the nomenclature suggested by Heywood (1964) and generally 
accepted in the recent pteridological literature (Page 1982; Lellinger 1985). 

l^In a taped interview with Nancy J. Turner, 1971. The ferns being discussed included 
Athyrium filix-femina, "Dryopteris austriaca [D. expansa], " Adiantum pedatum, and Blechum 
spicant, all of which George Young called, at times,djagwal.D. expansa was also called 
sndndjang; Pteridium aquilinum he called sndndjang-xil; and large specimens of Athyrium 
filix-femina, Dryopteris expansa, and Poly stichum munitum were called tsagwel, or skyaw 
(lit. "tail"), which he specified pertained to the larger, edible rootstocks. They are 
also sometimes called tanskyaw (lit. "black-bear's tail"). The smaller specimens of 
these species and Blechnum spicant, with rootstocks too small to eat, were sometimes 
called snal-djat (lit. "scabby-girl") (Florence Davidson and Willie Matthews, Masset 
speakers, in Nancy J. Turner unpublished notes, 1971). 

15 Margaret Siwallance used this term specifically to refer to the dead fronds of Pteridium 
(bracken fern), but recently she applied it to the dead fronds of any fern, or even 
dead leaves of any tree (Randy Bouchard, unpublished notes, 1974). 

16 Voucher specimens at Royal British Columbia Museum Herbarium: Turner 1698— 
V 127,877 (Athyrium filix-femina) and Turner 1699— V 127,878 (Dryopteris expansa). 

17 Voucher: Turner 1592— V 88,796, 88,798-9. 

18 The following unpublished notes made by Leslie M. J. Gottesfeld, and in her possession, 
are cited in the manuscript: Gottesfeld 1987, Gottesfeld 1988 (Gitksan). 

19 An additional fern to be found in the preferred habitat of ax, namely a mountain snow- 
bed or avalanche chute, is bracken (Pteridium aquilinum). This fern was distinguished 
by Jeff Harris Sr. and Billy Blackwater of Kispiox as being shoulder high and was 
called haba ba'a. Both haba ba'a and dutntx are said to have larger leaves than ax 




M 



Museum 



Herbarium (V), but all were identified or verified by Adolf Ceska, Jan. 30, 1989 as 
D. expansa: TM 201, TM 202, Eth 25, Eth 29, Eth 30. 

20 This information was mentioned elsewhere by Gordon Robinson in a handwritten 



file 



manuscript, transcribed from a recording of a speech by him on Haisla culture, on 
at the Native Indian Teacher Education Program Library, University of British Columbia 
Vancouver (no date given). 



terms 



These 



ine eaiDie wooa rem wac »«*»—- 

include terms meaning "soft end of 
edible fern root," "to eat edible fern root," "to go after edible fern root" (all Haisla), 
and "to eat fern roots" (Heiltsuk). There are also terms (Haisla, k'alak; Heiltsuk, k f alax> 
k'lk'ax) apparently referring to the fronds of wood fern, or "a long green fern species" 
(Lincoln and Rath 1986) which were used to cover salmon in a canoe to prevent them 
from drying out. 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 31 



22 



23 



Randy Bouchard, unpublished notes, 1973, 1975, British Columbia Indian Language 
Project, Victoria, British Columbia. 

slancy J. Turner (1973; unpublished notes, 1983) reported that Margaret Siwallace 
stated that people wishing to lose weight ate wood fern rootstocks raw, but Harriet 
Kuhnlein, who also spoke to MS about this, felt that she meant the rootstocks took 
so much energy to find and dig out that by the time you got enough for a meal, you 
would lose weight. Kari (1987) (see section, Survival and famine food) implies that it 
is possible to gain weight from eating (cooked) fern rootstocks. 



24 Th 



boiled fish, and certain greens such as sour dock, horsetail, mare's tail, or wood fern. 
In the winter it is stored in the cold so that the oils congeal, and it is served in a 
solid state. It is a favorite dessert at Napaskiak, and an adult usually will consume 



1957) 



25rh 



hours 



rw 



corn meal 



(25 



1987) 



26 Swanton noted "evidently fragmentary" for this myth. 
27 Hill-Tout (1905) identifies "Tsukwa" only as "some kin< 



grows 



ACKNOWLEDGEMENTS 



We would like to acknowledge the following aboriginal consultants for their contri- 
butions to this paper: Jeff Harris Sr., Solomon and Kathleen Marsden, Art Matthews Sr. 
and Kathleen Matthews, Alvin Weget, and Fern Stevens (Gitksan); Andrew George, 
Leonard George, Josephine Michell, Lizette Naziel, Kathryn Naziel, Sara Tait, Madeline 
Alfred, Alfred Joseph, and Elsie Tait (Wet'suwet'en); the late Dr. Margaret Siwallace and 
Felicity Walkus (Nuxalk); Gordon Robertson (Kitlope Haisla; interviewed by Brian Com- 
pton); Gordon and Phyllis Robinson and Lloyd Starr (Haisla); George Young (Haida); Mark 
Jacobs Jr. (Tlingit); Johnny Joe (Sechelt; interviewed by Randy Bouchard, 1978); Bill and 
Rose Mitchell, and Jeannie Dominick (Mainland Comox; interviewed by Randy Bouchard, 
1975). Many others provided information through published sources, and are acknowledged 
by name in most of these. Native language translators included Doris Michell and Cecilia 
Lapalme (Wet'suwet'en) and Beverley Anderson and Barb Senott (Gitksan). 

Samples for nutrient analyses were collected at Bella Coola by Sandy Burgess, Sarah 
Saunders, Dana Lepofsky, Aaron Hans, and David Hunt. 

We are also grateful to the following people for their help in providing information 
and/or editorial advice: Randy Bouchard and Dorothy Kennedy of the British Columbia 
Indian Language Project; Brian Compton; Dr. Eugene Hunn; Dr. David French; Dr. 
Barbara Lane; Dr. M. Dale Kinkade; Robert D. Turner; Dr. Richard Hebda; Robert Tyhurst; 
Beverley Anderson; and Allen Gottesfeld. Financial support to LMJG for various parts 
of this research was provided by the Gitksan-Wet'suwet'en Education Society, the 
Gitksan- Wet'suwet'en Traditional Medicine Project, the Kyah Wiget Education Society, 
and the Secretary of State. 



32 



TURNER 



Vol. 12, No. 1 



The National Museums 



Museum 



permission 



Museums 



Museum of Natural History, Ottawa (CAN), and Dr. K.M 
(Collections Manager), are acknowledged for the loan of Harlan Smith 



The staff of the University of Washington herbarium (WTU) 



Museum 



(V) 



LITERATURE CITED 



BARBEAU, MARIUS. 1929. Totem poles of 
the Gitksan, Upper Skeena River, British 

Columbia. National Museum of Canada, 
Bulletin No. 61, Anthropological Series 
No. 12, Ottawa. 

BOAS, FRANZ. 1921. Ethnology of the 
Kwakiutl. Smithsonian Institution, The 
Bureau of American Ethnology, 35th 
Annual Report, Parts 1 and 2, 1913-14. 

1928. Bella Bella Texts. Colum- 



bia Univeristy Press, New York. 
1932. Current beliefs of the 



Kwakiutl Indians. lournal of American 
Folklore 45:177-261. 

1934. Geographical Names of 

the Kwakiutl Indians. Columbia Univer- 
sity Press, New York. 

BOUCHARD, RANDY AND NANCY J. 
TURNER. 1976. Ethnobotany of the 
Squamish Indian People of British 
Columbia. Report to the Squamish 
Indian Band, North Vancouver, British 
Columbia. British Columbia Indian Lang- 
uage Project, Victoria. 

BRTTTON, DONALD M. 1962. Dryopteris 
dilatata (Hoffm.) A. Gray in North Amer- 
ica. Rhodora 64 : 207-212 . 

1968. The spores of four 

species of spinulose woods ferns (Dryop- 
teris) in eastern North America . Rhodora 



70:340-347. 



ornamentation 



the Dryopteris spinulosa complex. 
Canadian Journal of Botany 50:1617- 
1621. 

CALDER, JAMES A. AND ROY L. TAYLOR. 
1968. Flora of the Queen Charlotte 
Islands, Parts 1 and 2. Canada Depart- 
ment of Agriculture, Research Branch, 
Monograph No. 4, Ottawa. 

CLAUS, EDWARD P., VARRO E. TYLER, 
AND LYNN R. BRADY. 1970. Pharma- 
cognosy. Lea & Febiger, Philadelphia. 



RTIS, EDWARD S. 1907. The Haisla. 
In The North American Indian, Vol. 3. 
The University Press, Norwood, Massa- 
chusetts (Reprinted in 1970 by Johnson 
Reprint Corporation, New York.) 

. 1915. The Kwakiutl (Southern 

Kwakiutl, Oowekyala). In The North 
American Indian, Vol. 10. The University 
Press, Norwood, Massachusetts. (Re- 
printed in 1970 by Johnson Reprint Cor- 
poration, New York.) 
. 1916. Nootka; Haida, pp. ix- 




235 In The North American Indian, voi. 
11. The University Press, Norwood, Mas- 
sachusetts. (Reprinted in 1970 by John- 
son Reprint Corporation, New York.) 
DUNN, JOHN A. 1978. A practical dictionary 
of the Coast Tsimshian language. 
National Museum of Man, Mercury 
Series, Canadian Ethnology Service 

Paper No. 42, Ottawa. 

GALLOWAY, BRENT. 1982. Upriver Hal- 
qemeylem Ethnobotany. Coqualeetza 
Education Training Centre, Sardis, 
British Columbia. 

GITANYOW SUMMER STUDENT 

RESEARCH PROGRAM. 1984. Report 
on Traditional Foods and Medicines, 
Phase II. Medical Services, Job Creation 
Branch, Ottawa, Ontario. 

GORMAN, M.W. 1896. Economic botany or 
S.E. Alaska. Pittonia 3(14):64-85. 

GUNTHER, ERNA. 1973. Ethnobotany or 
Western Washington. Revised Edition. 
University of Washington Press, Seattle. 

HELLER, CHRISTINE. 1976. Wild ediDie 
and poisonous plants of Alaska. °- 
operative Extension Service Pubhcatio 
No. 28, University of Alaska, College- 

HEYWOOD, V.H. 1964. Dryopteris Adanson. 
Pp. 20-22 In Tutin, T.G. et al. (editors). 
Flora Europaea, Vol. 1: Lycopodiaceae 
to Platanaceae. Cambridge University 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



33 



LITERATURE CITED (continued) 



Press, Cambridge. 

HILL-TOUT, CHARLES. 1905. Report on the 
ethnology of the Stlatlumh [Lillooet] of 
British Columbia. Journal of the Anthro- 
pological Institute of Great Britain and 
Ireland 35:126-218. 

HULTEN, ERIC. 1968. Flora of Alaska and 
Neighboring Territories. Stanford Uni- 
versity Press, Stanford. 

JACOBS, MARK, JR. AND MARK JACOBS, 
SR. 1982. Southeast Alaska native foods, 
Pp. 112-130 In Raven's Bones. Andrew 
Hope (editor). Sitka, Alaska. 

KARI, PRISCILLA R. 1987. Tanaina Plant- 
lore, Dena'ina K'et'una: An Ethnobotany 
of the Dena'ina Indians of Southcentral 
Alaska. National Parks Service, Alaska 
Region, Anchorage. 

KINKADE, M. DALE. 1989. Comparative 
Linguistic Evidence About Salish Pre- 
history. Paper presented to American 
Anthropological Association, Washing- 
ton, D.C. 

'KSAN, PEOPLE OF. 1980. Gathering What 
the Great Nature Provided: Food Tradi- 
tions of the Gitksan. Douglas and 
Mclntyre, Vancouver, British Columbia. 

KUHNLEIN, HARRIET V. 1989a (in press). 
Change in the use of traditional foods 
by the Nuxalk Native people of British 
Columbia. In Perspectives of Dietary 
Change: Studies in Nutrition and Society 
G.H. Pelto and L.A. Vargas (editors). 
International Nutrition Foundation for 
Developing Countries, Cambridge, Mas- 
sachusetts. 

1989b. Nutrient values in 



indigenous wild berries used by the 
Nuxalk of Bella Coola, British Columbia. 
Journal of Food Composition and Anal- 
yses 2:28-36. 

__^ , NANCY J. TURNER, AND 

PAUL D. KLUCKNER. 1982. Nutritional 
significance of two important root foods 
(springbank clover and Pacific silver- 
weed) used by Native people on the coast 
of British Columbia. Ecology of Food and 
Nutrition 12:89-95. 

LELLINGER, D.B. 1985. A Field Manual of 
the Ferns & Fern Allies of the United 
States & Canada. Smithsonian Institution 
Press, Washington, D.C. 



LEPOFSKY, DANA, NANCY J. TURNER, 

AND HARRIET V. KUHNLEIN. 1985. 

Determining the availability of traditional 

wild plant foods: An example of Nuxalk 

foods, Bella Coola, B.C. Ecology of Food 
and Nutrition 16:223-241. 

LINCOLN, NEVILLE J. AND JOHN C. 
RATH. 1986. Phonology, dictionary and 
listing of roots and lexical derivates of 
the Haisla language of Kitlope and Kiti- 
maat, B.C. (2 Vols.). Canadian Museum 
of Civilization, Mercury Series, Canadian 
Ethnology Service Paper No. 103, 
Ottawa. 

McILWRAITH, THOMAS F. 1948. The Bella 
Coola Indians, Two Volumes. University 
of Toronto Press, Toronto. 

McNEARY, STEVEN. 1974. Article on Niska 
plant use. Report to National Museum of 
Canada, Ottawa. 

MORICE, ADRIAN G. 1893. Notes archae- 
ological, industrial and sociological on 
the Western Denes. Transactions of the 



222 



HELEN 



bracken as food for aboriginal peoples 
of western Washington. Economic 
Botany 33:384-396. 

. 1981. Plant use in Kaigani 

Haida culture: Correction of an ethno- 



Economic 



449 



OSWALT, W.H. 1957. A western Eskimo 



Anthropological 



36 



PAGE, C.N. 1982. The Ferns of Britain and 
Ireland. Cambridge University Press, 



Cambridge. 



AND 



1973. Additions to the Quileute Entries, 



Western 



Washington by Erna Gunther. University 
of Washington Press, Seattle. 
t 9 GLEN. 1980. Root, Stem and Leaf: 
Wild Vegetables of Southeast Alaska. 
South East Regional Resource Center, 

Juneau. 

DGGAN, H.J. 1978. The Flora of Canada, 

Part 2. National Museum of Natural 

Sciences, National Museums of Canada, 

Ottawa. 

IALEY, GERALD B., ROY L. TAYLOR, 



34 



TURNER et al. 



Vol. 12, No. 1 



LITERATURE CITED (continued) 



AND GEORGE W. DOUGLAS. 1985. 
The rare vascular plants of British 
Columbia. Syllogeus 59:1-165. 

SMITH, HARLAN I. 1928. Materia medica 
of the Bella Coola and neighbouring 
tribes of British Columbia, Pp. 47-68 
In National Museum of Canada, 1927 
Annual Report, Ottawa. 

SW ANTON, IOHN R. 1909. Myths and texts 
of the Tlingit. Smithsonian Institution, 
Bureau of American Ethnology Bulletin 

No. 39. 
TAYLOR, ROY L . AND BRUCE 

MACBRYDE. 1977. Vascular Plants of 
British Columbia. University of British 
Columbia Press, Vancouver. 

TEH, IAMES. 1909. The Shuswap. Vol. 2, 
Part 7. The Jesup North Pacific Expedi- 
tion, Franz Boas (editor). American 
Museum of Natural History, New York. 

TURNER, NANCY 1. 1973. The ethnobotany 
of the Bella Coola Indians of British 
Columbia. Syesis 6:193-220. 

. 1975. Food plants of British 



Columbia Indians, Part 1: Coastal 
peoples. British Columbia Provincial 
Museum, Handbook No. 34, Victoria. 
AND MARCUS A.M. BELL. 



The 



Kwakiutl Indians of British Columbia. 
Economic Botany 27:257-310. 

AND BARBARA S. EFRAT. 

1982. Ethnobotany of the Hesquiat 
Indians of Vancouver Island. British 
Columbia Provincial Museum, Cultural 



Recovery Paper No. 2, Victoria. 



THOMAS 



OGILVE 



1983. Ethnobotany of the Nitinaht 
Indians of Vancouver Island. British 
Columbia Provincial Museum, Occa- 
sional Paper No. 24, Victoria. 

. LAURENCE C. THOMPSON, 



M. TERRY THOMPSON, AND ANNIfc 
Z. YORK. 1990. Thompson Ethnobotany: 
Knowledge and Usage of Plants by the 
Thompson Indians of British Columbia. 



Museum 



Memoir No. 3, Victoria. 



WALKER, W 



Cytogenetic studies in 



the Dryopteris spinulosa complex I. 



Watsonia 3:193-209. 

. 1961. Cytogenetic studies in 

the Dryopteris spinulosa complex II. 
American Journal of Botany 48:607-614. 

WATT, BERNICE K. AND ANNABEL L. 
MERRILL. 1975. Handbook of the Nutri- 
tional Contents of Foods. Dover Publica- 
tions, New York, Revised Edition, (origi- 
nally published in 1963 as United States 
Department of Agriculture Handbook 
No. 8, Composition of Foods. Consumer 
and Food Economics Research Division 
of USDA, Washington, D.C. 

WIDEN, C.J. AND DONALD M. BRITTON. 
1971 . A chromatographic and cytological 
study of Dryopteris dilatata in North 
America and eastern Asia. Canadian 
Journal of Botany 49:247-258. 






BOOK REVIEW 



Conservation of Medicinal Plants: Proceedings of an International Cons 
tation. Olayiwola Akerele, Vernon Heywood, and Hugh Synge (editc > /. 
Cambridge, United Kingdom: Cambridge University Press, 1991. Pp. xvi, do • 
No price given (hardcover). ISBN 0-521-39206-3. 

This impressive volume reports on the findings of an international _ m ^ tir ]^ 
held in Chiang Mai, Thailand, from 21-27 March, 1988. Sponsored by the wo 
Health Organization (WHO), the World Conservation Union (IUCN), 
World Wide Fund for Nature (WWF), it brought together experts from 16 cou ^ on 
to "exchange views on the problems, determine priorities and make r 
dations for action" regarding the conservation of medicinal plants. 



and the 



ecommen 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 35 



ustomary 



Chiang Mai 



nternational 



social and ecological deterioration that threaten plants used in health care. 

The book is separated into six sections. The Introduction reviews the perspec- 
tive on medicial plants taken by the sponsoring agencies. The Issue of Medicinal 
Plants contains papers by Farnsworth and Soejarto, Plotkin, and Schultes. 
Although these overviews provide a good introduction for novices, most of the 
programs, concepts, and references will be familiar to practicing ethnobotanists. 

The section on Science, Industry, and Medicinal Plants begins with three 
technical papers that analyze economic aspects of plant conservation and manage- 
ment. A fourth paper describes recent advances in constructing information 
systems and databases. Written by Synge and Heywood, two leaders in the field, 
it contains useful suggestions on how to use computers to arrange research data. 

The remaining three sections present the applied side of medicinal plant 
conservation-Techniques, Policies, and Experiences. Eighteen papers review 
diverse topics ranging from the role of agronomy, botanical gardens, and pro- 
tected areas to political, legal, and educational aspects. Case studies focus primarily 



Thailand 



m 



dichotomy 



between in situ and ex situ conservation. Although both will ultimately be 
needed in any campaign to preserve med 



camp and the other. Ex situ conservation is 
governmental or international agency. Some 
biotechnology (Palevitch; Schumacher), 
botanical gardens (Heywood; He and Cheng), phytochemical research (Husain), 
and commercialization (Principe; Bonati)— take the resources out of the hands 



method 



producers 



property rights (IRP) 



information system 



commercialization 



among the case studies from the Third World. This may send the message that 
computer networking and information sharing will ultimately benefit industrializ- 
ed countries, while giving little profit to the communities from which plants and 
ethnobotanical lore are gleaned. In sum, this set of papers would have been enrich- 
ed by a greater emphasis on how technology can help indigenous peoples and 
developing nations maintain a stake in their cultural heritage and natural 

resources. 

With the essay by McNeely and Thorsell on the role of protected areas in 

conserving medicinals, there is a shift in emphasis: peasants and indigenous 
peoples are back in the equation. The case study on in situ conservation in Sri 



demonstrates how medicinal plants 



com 



panion essay (de Alwis) shows how ex situ conservation supports th 

The volume is fittingly brought to a close with a case study from < 

(Cunningham) that exemplifies how theory and methodology from 



36 



BOOK REVIEW Vol 12, No. 1 



resource management can be applied to conservation of plant populations. This 
approach gives us the tools necessary to evaluate the status of medicinal plants 
in the wild and to propose alternative management methods that are acceptable 
to local people. 

When I closed the book after reading the parting words of the WHO pro- 
gram manager for traditional medicine, Olayiwola Akerele, I found myself reflec- 
ting on the speed with which things are changing. In the three and a half years 
since the meeting was convened, the agenda advocated by the conference parti- 
cipants has been widely accepted and implemented in some countries. At the 
same time, the world ecological and social situation is increasingly dire. I recom- 
mend this volume to anyone who wants to join the battle for indigenous rights 



and natural conservation. 



Gary J. Martin 
Anthroplogy Department 
University of California 
Berkeley, CA 94720 



/. Ethnobiol. 12(l):37-62 



Summer 1992 



INFLUENCES OF MID-HOLOCENE ALTITHERMAL 

CLIMATES ON MAMMALIAN FAUNAS 
AND HUMAN SUBSISTENCE IN EASTERN WASHINGTON 



R. LEE LYMAN 
Depa rtmen t of Ant h ropology 

200 Swallow Hall 

University of Missouri-Columbia 

Columbia, Missouri 65211 



ABSTRACT.— Palynological data indicate climates in eastern Washington between 
8000 and 4000 B.P. were warmer and drier than before or after that time. It has 
been hypothesized that this mid-Holocene warm-dry interval, typically called the 
Altithermal, would have resulted in decreased mammalian biomass and prompted 
prehistoric hunter-gatherers in eastern Washington to shift subsistence pursuits 
to focusing on fish and plants, increasing reliance on small mammals, a broader 
range of mammalian taxa being exploited, or some combination of these. Mam- 
malian faunal data compiled from 11 sites in one area and 28 sites in another area 



conform 



000 



areas, fish remains are more abundant after 4000 B.P. than before that time in 



000 



in both areas. Indications of natural faunal turnover are masked by sample size 
effects, and shifts in mammalian biomass are obscured by varying intensity of 
human occupation. 



RESUMEN.— Los datos palinologicos indican que, entre 8000 y 4000 anos antes 
del presente, el clima en el este del Estado de Washington (Estados Unidos de 
Norteamerica) fue mas calido y mas seco que antes o despues de dicho tiempo. 
Se ha aventurado la hipotesis de que este intervalo caliente y seco de mediados 
del holoceno, llamado tipicamente el altitermal, habna resultado en una biomasa 
reducida de mamiferos y habna provocado que los cazadores-recolectores en el 
este de Washington modificaran su subsistencia para enfocarse en peces y plan- 
tas, incrementar su dependencia de mamiferos pequenos, explotar un rango mas 
amplio de taxa de mamiferos, o alguna combinacion de estas posibilidades. Los 
datos compilados sobre mamiferos de 11 sitios en un area y 28 sitios en otra, sin 
embargo, no concuerdan con estas hipotesis. La abundancia relativa de mamiferos 
pequenos usados como alimento disminuye continuamente durante los ultimos 
10,000 anos en ambas areas, los restos de peces son mas abundantes despues 
de 4,000 anos antes del presente en un £rea, y los ungulados se vuelven mas 
y ma's abundantes a lo largo de los ultimos 10,000 anos en ambas areas. Las 
evidencias de la rotacion natural de existencias de fauna son disfrazadas por 
efectos del tamano de las muestras, y los cambios en la biomasa de mamiferos 
son oscurecidos por la intensidad variable de la presencia humana. 



38 



LYMAN Vol. 12, No. 1 



* * 



RESUME.— Les donnees palynologiques indiquent que les climats dans l'Est de 
Washington, entre 8000 et 4000 B.P. etaient plos chauds et sees qu'arant et opres 
cette periode. On a propose que cet interval chaud et sec du Moyen Holocene, 
rAltiehermal, aurait pu declancher un declinement de la faune mammifere. En 
consequence, les chasseurs collecteurs prehistoriques de l'Est de Washington 
auraient modifier leur existence en diminvant l'importance des grands mammifers 
et en accentuant l'importance du poisson, des plantes, et des petits mammifers, 
ou un plus grand rayon d'exploitation de types de mammiferes, ou pent-etre une 
combination de ceux-ci. Les donne'es recueillies peur la faure mammifere de 11 
sites dans une region et 28 sites d'une autre region ne supportent pos ces 
hypotheses. Les petits mammiferes utilizes comme denree alimentaire diminuent 
continuellement en abendance relative pendant les dernieres 10,000 annees pour 
les 2 regions. Les restes de poissons sont plus abondants apres 4000 B.P. dans 
une region, et les ongule's augmentent constament pendant les dernieres 10,000 
annees dans les 2 regions. Les indications de renversements de faune naturels 
sont masquees par les petites dimension des groupes etudies, et les changements 
de faune mammifere sont voile's par les variations dans l'intensite d'occupation 
humaine. 



INTRODUCTION 



Palynological data for Holocene deposits in eastern Washington (summarized 
in Barnosky et. al. 1987; Mehringer 1985) suggest that climates 10,000 years ago 
were cooler and moister than at present. Evidence of increased aridity appears 
as early as 9000 B.P. or slightly later. Maximal aridity seems to have occurred 
between approximately 8000 B.P. and 5000 to 4500 B.P. during the climatic inter- 
val called the Altithermal, with essentially cooler and moister to modern condi- 
tions prevailing after 4000 B.P. Potential impacts of the mid-Holocene warm, dry 
climatic interval on human occupants of the area have long been the subject of 
discussion (e.g.,'Baumhoff and Heizer 1965; Bense 1972). Fryxell and Daugherty 
(1963:14) suggested large game would decrease in abundance and force prehistoric 
peoples to rely on fishing and gathering rather than hunting during the mid- 
Holocene. Schalk (1983:145) argued xeric climatic conditions would decrease 
ungulate biomass due to decreased biomass of forage and "would probably 
exclude ungulates other than pronghorn (Antilocapra americana)" from the more 
xeric areas of eastern Washington (see Van Vuren 1987 for similar arguments 
concerning bison [Bison bison]). 

Previous studies of the mammalian history of eastern Washington indicate 
some taxa apparently responded to climatic change during the Holocene by modi- 
fying their distributions (Lyman 1986a, 1986b, 1991a, 1991b; Lyman and Livingston 



McCorquodale 



Holocene 



Research at Marmes Rockshelter (site number 45FR50) suggested to Gustafson 
(1972:105) that mammalian faunas in eastern Washington remained taxonomically 
stable throughout the last 8000 years. He detected an increase in the abundance 
of large mammals during the Altithermal occupation of Marmes Rockshelter which 
he felt resulted from the fact that the faunal remains were humanly deposited 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 39 






and thus reflected both the number of animals killed for human consumption 
rather than natural biomass, and a mid-Holocene increase in human occupational 
intensity of the site. In this paper I review the known Holocene record of 
mammals in two geographically distinct areas of eastern Washington in an 
attempt to find evidence of changes in the mammalian fauna and the contribu- 
tion of that fauna to human subsistence. Simultaneously I search for sample-size 
effects as it has been well documented that sample size can adversely influence 
the measurement of faunal turnover (e.g., Badgley and Gingerich 1988; Koch 1987) 
and estimates of faunal biomass (e.g., Grayson 1984). I also monitor the inten- 
sity of human occupation because increases in such occupation could result in 
higher rates of deposition of animal remains in sites. 



STUDY AREA, MATERIALS, AND METHODS 












The Columbia Basin of eastern Washington is diverse physiographically and 
vegetationally (Franklin and Dyrness 1973) and contains several vegetation zones 
that Daubenmire (1970) has characterized generally as steppe. Mean annual 
precipatation is 20 cm in the southern (Lower Snake River) part of the basin and 
33 cm in the northern (Upper-Middle Columbia River) part (Fig. 1). Mean 
temperature in January is -5° C in the southern part and -9° C in the northern 
part; mean temperature in July is about 32° C in the southern part and 30° C 
in the northern part. 

To date, few data on the Holocene mammals of eastern Washington have 
been derived from natural bone accumulations such as natural traps, carnivore 
dens, or raptor or owl roosts. The single such studied fauna was recovered from 
deflated areas in a dune field and is of unclear but probably late Holocene age 
(Miller 1977). Well over 95% of the available mammalian data derive from 
archaeological sites that have been excavated by numerous individuals over the 
past 35 years. Because humans were responsible for selectively accumulating and 
depositing many of the bones in such sites, care is warranted when attempting 
to derive paleoenvironmental meaning from these bone assemblages. The sites 
I have used in this analysis are, with one exception (Marmes Rockshelter), open 
sites, but some are villages (i.e., have clear evidence of substantial structures) 
while others appear to be camps (i.e., lack evidence of structures), suggesting 
between-site variation in season of occupation, duration of occupation, or both. 
Assemblages of artifacts and plant remains indicate that activities performed by 
human occupants of the sites were similar but not identical, with some between- 
site variation in resources procured, procurement techniques used, and resource 
processing. Geological data indicate between-site variation in postdepositional 
histories of the faunal remains. On one hand, these data indicate that while 
interpretation of the mammalian fauna from a single site might be significantly 
influenced by the human activities and postdepositional proccesses that, respec- 
tively, produced and affected the faunal remains, summing the faunas from a 
series of sites should serve to mute such influences and produce a general 
indication of trends in the natural faunal history of the area (Lyman 1987). On 
the other hand, the muting effects of summing multiple archaeological faunas 
will produce only general indications of trends in human subsistence pursuits, 



40 



LYMAN 



Vol. 12, No. 1 



49 



48 



47 



46 



49° N 




the Columbia Basin (circumscribed by the Columbia 



River area in eastern Washingt 



River area, and the Upper-Middle 



some sites seem to have been occupied bv humans 



more 



specialized in subsistence pursuits than the occupants of other sites. This seems 
preferable to site-specific studies at this time; once general trends are established 
in both the cultural or subsistence pursuits of people and the natural faunal history 
of the area, site-specific faunas from both cultural and natural deposits can be 
used to fine-tune the trends indicated. 

Data were compiled from published and unpublished archaeological reports 
for two areas: the western two-thirds of the Lower Snake River, and the Wells 
Dam Reservoir and Chief Joseph Dam Reservoir areas of the Upper-Middle 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



41 









Columbia River (Fig. 1). These two areas were chosen for study because the former 
is the geographic location from which data leading to discussions of Altithermal- 
induced faunal change originated, and the latter area contains the largest sample 
of mammalian remains of any area in eastern Washington similar in geographic 
size to the Lower Snake River area. 

Mammalian faunal data from 11 archaeological sites along the lower Snake 
River and 28 archaeological sites along the Upper-Middle Columbia River were 
compiled (Tables 1 and 2). Data from the former 11 sites span the last 10,000 years 
while data from the latter 28 sites span the last 7000 years. Well over half of the 
Upper-Middle Columbia River data derive from the Chief Joseph Dam Reservoir 
area. These data were originally reported by three temporal periods: 7000 to 4000 
B.P., 4000 to 2000 B.P., and 2000 to 100 B.P. (see Livingston 1985 for a summary 
of the data). Thus I tallied all other faunal data for this area by those three 
temporal periods despite the fact that many of these data had been assigned to 
much briefer temporal spans (Chatters 1984a, 1984b, 1986; Lyman 1988). I tallied 
the Lower Snake River data by 2000 year increments for the entire Holocene to 
make them readily comparable to those from the Upper-Middle Columbia River 
area. While such arbitrary definitions of temporal periods may mask significant 
turnover events or instances of fluctuating biomass (e.g., Peterson 1977; Schindel 
1980), by lumping site-specific samples I hoped to eliminate sample size-effects 
such as larger samples typically being taxonomically richer than smaller samples. 



TABLE 1.— Sample size (number of identified specimens [NISP]) per time interval 
and site in the Lower Snake River area. 






Site 



45FR50 
45WT2 
45CO! 






45 G A6 1 

45WT39 

45GA17 

45WT41 

45FR36 

45FR5 



45FR40 

45WT134 



Years B.P. 



10,000- 8000- 6000- 4000- 2000- 



8000 



6000 



4000 



289 



269 
42 
44 



158 
17 



324 



41 



2000 



57 
17 




571 
161 

53 



283.5 



100 



57 
38 



648 
808 



65 



4,749 



35 



Reference 



Caulk 1988; Gustafson 1972 

Gustafson 1972 

Bense 1972; 
Nelson et al. 1968 

Lyman, unpubl.l 

Yent 1976 

Lyman, unpubl.l 

Gustafson 1972 

Schalk 1983 

Schalk and Olson 1983; 
Olson 1983 

Kenaston 1966 



296.5 Lyman 1990 






1 



Unpublished lab data in possession of the author. 



42 



LYMAN 



Vol. 12, No. 1 



TABLE 2— Sample size (number of identified specimens [NISP]) per temporal 
interval and site in the Upper-Middle Columbia River area. 



Site 



Years B.P. 



7000- 
4000 



45D0189 

45DO190/191 

45DO204 

45D0211 

45D0214 

45D0242 

45D0243 

45D0273 

45D0282 

45D0285 

45D0326 

45D0372 

45D0394 

450K2 

450K2A 

450K4 

450K11 

450K18 

450K69 

450K165 

450K197 

45OK207 

45OK208 

45OK250 

450K258 

450K287/288 

450K382 

450K383 



31 



53 

48 

426 



283 



45 

131 

3,034 



105 

68 

213 



182 
50 



4000- 

2000 



452 
109 



479 

189 

609 

94 



326 

79 
496 

31 
661 
294 
790 
540 

31 
104 

31 



666 

2,332 

451 



241 



2000- 
100 



400 
50 



196 

276 

42 



1401 



97 



1253 



1,986 
200 



Reference 



Lyman 1988 
Chatters 1984b 
Livingston 1985 
Livingston 1985 
Livingston 1985 
Livingston 1985 
Livingston 1985 
Livingston 1985 
Livingston 1985 
Livingston 1985 
Livingston 1985 
Chatters 1986 
Chatters 1984b 
Livingston 1985 
Livingston 1985 
Livingston 1985 
Livingston 1985 
Livingston 1985 
Chatters 1986 
Chatters 1984b 
Chatters 1984b 
Chatters 1984b 
Chatters 1984b 
Livingston 1985 
Livingston 1985 
Livingston 1985 
Chatters 1986 
Chatters 1986 



Taxonomic abundances were tallied as numbers of identified specimens 
(NISP). While there are sound theoretical reasons for using NISP rather than a 
different quantitative unit in my analysis (Grayson 1984), the practical reason is 
that quantitative data are only regularly reported in the form of NISP for the 
materials I analyze. It is difficult to estimate biomass directly from NISP values, 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 43 



but traditional interpretations in the geographic area under consideration have 
used NISP as a fiat measure of animal biomass, and I follow that tradition here. 

For both areas only temporally fine-scale assemblages that include more than 
30 NISP were used to remove some of the effects of excessively small samples. 
The total NISP from a site might be 59, but if those specimens were strati- 
graphically distributed in such a manner that 28 specimens dated between 2000 
and 1500 B.P. and 31 specimens dated between 3000 and 2000 B.P., generally 
only the latter 31 specimens were included in my analysis. By summing site- 
specific data within each area, I hoped to compile an area-specific faunal history 
for both areas and to avoid the effects of site-specific depositional hiatuses and 
temporally fine-scale but small assemblages on my analyses. In a few cases site- 
specific faunal assemblages span time periods greater than 2000 yr or overlap a 
temporal boundary. For example, one stratigraphically defined faunal assemblage 
was dated between 5000 and 3000 B.P. This assemblage was simply halved, with 
exactly half the NISP per taxon being tallied for the 6000 to 4000 B.P. period and 
the other half being tallied for the 4000 to 2000 B.P. period. Only one strati- 
graphically defined fauna (45WT2) that was split into two faunas, each with fewer 
than 30 NISP, was used in my analysis (Table 1). 

I noted the temporal occurrence of each taxon represented at each site in both 

areas and compiled two tables, one for each geographic area (Tables 3 and 4). 

A taxon was deemed present for an entire temporal period if its remains dated 

to any part of that period. I then tallied the total NISP per taxon per area, the 

number of sites in which a taxon was found per area, the taxonomic richness 

per temporal period per area, and the total NISP per temporal period per area. 

Because zooarchaeological data are often at best ordinal scale (Grayson 1984), I 

use Spearman's rho (r s ) to assess the strength of correlations between selected 
pairs of variables. 

Measures of human occupational intensity such as depositional rates of 
artifacts are only available for some of the sites I consider. I use such measures 
when available, and also use frequencies of radiocarbon dates and frequencies 
of dated sites as measures of occupational intensity. I fully realize such fre- 
quencies are at best indirect measures of human occupational intensity because 
frequencies of dates and dated sites often reflect an archaeologist's research design 
or preservation of dateable materials (Rick 1987). I therefore supplement these 
data when possible with information on sites dated by the presence of temporally 



diagnostic artifacts. 



HYPOTHESES AND IMPLICATIONS 



Palynological data alone lead to the following hypotheses: 

Animal biomass should be at its lowest level during the 
eastern Washington. 



Ungulates such as bison, deer (Odocoileus spp.), wapiti (Cervus elaphus), and 
bighorn sheep (Ovis canadensis) should display their lowest abundance (and 
may not be present) during the mid-Holocene. 



44 



LYMAN Vol. 12, No. 1 



3) The number of taxa and the frequency of individual taxa, particularly those 
adapted to relatively moist conditions, should both be at their lowest levels 
during the mid-Holocene. 

4) Changes in taxonomic composition through time should be contemporaneous 
with changes in climate. However, as noted, available faunal data are from 
archaeological contexts wherein the human selection of prey may significantly 
influence the faunal record. It is important, therefore, to consider the possible 
combined influences of human behaviors and increased aridity on faunal 
abundances. 



Human responses to climatic change. —Resources can become scarce through an 
actual decrease in frequency or through the effect of increasing competition for 
resources as the relative frequency of predators/consumers increases. Ecologists 
believe that niche breadth tends to increase as resource availability decreases 
(Pianka 1978:256) and that when high-value prey are rare, diet expands to 
include more low-value prey (Pulliam 1981:65). Anthropological research indicates 
human hunter-gatherers tend to respond in such manners (e.g., Belovsky 1987, 
1988; Bettinger 1991; Colson 1979; Earle 1980; Keene 1985; Hayden 1981; O'Con- 
nell and Hawkes 1981; Simms 1987; Winterhalder 1986) . Thus, small mammal 
resources should be most abundant relative to large mammal resources during 
the mid-Holocene when the latter were supposedly depleted. Secondly, the diver- 
sity of mammalian resources should be highest during the mid-Holocene as all 
resources were exploited with more or less equal intensity rather than a selected 
few. 

The effect of increased intensity of human occupation on hypotheses (1), (2), 
and (3) is for those hypotheses to have reverse implications; that is, frequencies 
of animal remains and taxonomic richness should increase as people shift adap- 
tive strategies in response to increased aridity despite the fact that naturally 
available mammalian biomass may be decreasing in abundance. As dietary breadth 
increases, number of taxa (richness) :NISP ratios should increase (e.g., from 1:10 
to 2:10). If there are also indications that fewer sites were occupied during the 
mid-Holocene, and that those sites that were occupied were more intensively 
occupied during the mid-Holocene than before or after that time, then, given the 
argument presented here, these kinds of changes in these ratios can be taken 
as evidence for a decrease in naturally available biomass. However, if sites were 
occupied with a lower level of intensity during the mid-Holocene than before or 
after that time, the ratios described above may be similar between time periods 
if animal biomass was relatively high during the early and late Holocene and low 
during the mid-Holocene. That is so because lower occupational intensity sug- 
gests a lower level of competition than does high occupational intensity. 

Faunal turnover should be indicated by discontinuous or incomplete temporal 
representation of taxa through the Holocene. However, if sample size, measure 
as the number of sites producing remains of a taxon or measured as the Nib 
per taxon, is influencing the apparent temporal range over which a taxon is 
found, then I expect to find statistically significant positive correlations between 
such variable pairs as taxonomic richness per temporal increment and NISP p e 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 45 



temporal increment, NISP per taxon per temporal increment and number 
temporal increments in which a taxon is found, and number of sites produci 
remains of a taxon and number of temporal increments in which a taxon is four 
Significant negative correlations between the number of gaps in the tempo 
distribution of a taxon and NISP per taxon are also expected if the available fau 
data tend to monitor sample size rather than faunal turnover. 



Faunal responses to drought. —Van Vuren (1987: see also Van Vuren 1984; Van Vuren 
and Bray 1983, 1985, 1986) suggests that the amount and distribution of herbaceous 
vegetation were the most significant factors controlling the abundance of bison 
during the Holocene of eastern Washington. Decreases in abundance of herba- 
ceous vegetation, increased space between patches of herbaceous vegetation, or 
both, would result in a decrease in the population of bison and perhaps their 
local extirpation. McCorquodale et al. (1986) report that wapiti typically do not 
occupy arid areas void of extensive thermal cover (typically forest) due to the high 
cost of thermoregulation. They indicate, however, that wapiti can occupy big 
sagebrush (Artemisia tridentata) steppe habitats where adequate forage is available 
and they are not frequently disturbed. Increased hunting pressure brought on 
by decreases in other large mammals might, then, prompt wapiti to abandon the 
sage-steppe vegetation zone. The size and distribution of populations of bighorn 
sheep in southeastern Washington and southern British Columbia today seem 
closely related to forage quality (Estes 1979; Pitt and Wikeem 1978). Drought- 
induced reduction in forage quantity and quality has been shown to correlate with 
higher than normal lamb mortality (Johnson 1983), and thus populations probably 
would decrease significantly if arid conditions were to persist over many years. 
Pronghorn do not require free-standing water when forbs are abundant and have 
high moisture content, but during dry periods water consumption by pronghorn 
is high (Beale and Smith 1970). When drought reduces the quality of forage, 
pronghorn populations decline (Kitchen and O'Gara 1982). 

Deer are polytypic in eastern Washington (both Odocoileus hemionus and 
O. virginianus are present), and while probably reduced in abundance during 
prolonged periods of aridity, this taxon would perhaps not have completely disap- 
peared from that region during the mid-Holocene. Large carnivores such as bears 
(Ursus spp.) and cougars (Felis concolor) are so rare in archaeological assemblages 
from eastern Washington that it is difficult to determine if their absence is due 
to environmental conditions or to the fact that they were rarely taken by prehistoric 
hunters (e.g., Lyman 1986b). When their remains are recovered from archae- 
ological contexts, they may thus be rather unreliable indicators of faunal change. 
At least some small mammalian taxa, such as the pygmy rabbit (Brachylagus 
idahoensis), seem to have increased in abundance and range during the mid- 
Holocene, while other small taxa, such as the pocket gopher (Thomomys talpoides) 
seem to have decreased in abundance at that time (Lyman 1991a). 

Finally, if mammalian herbivores decreased in abundance during the mid- 
Holocene Altithermal climatic interval, it seems likely that the frequency of non- 
human mammalian carnivores would have similarly decreased. That is so because 
of lowered food supplies for the latter taxa, some of which feed not only on prey 
they have hunted and killed, but also on carrion. However, because mammalian 



46 



LYMAN Vol. 12, No. 1 



carnivores tend to occur high in trophic levels, they are naturally rare 
interpretations of prehistoric changes in their abundance must be viewed 
caution. 



RESULTS AND DISCUSSION 

Lower Snake River. —Intensity of occupation of sites in the Lower Snake River area 
is difficult to assess because the volume excavated per stratigraphic unit has 
seldom been reported. Interestingly, the single site that spans the entire Holocene 
(Marmes Rockshelter) produced a series of radiocarbon dates that has been 
interpreted by Sheppard et al. (1987) as indicating minimal occupation of that 
site between 6700 and 1900 B.P. I compiled all radiocarbon dates from sites in 
the Lower Snake River area known to me. Ten of the eleven sites producing faunal 
data have associated radiocarbon dates (45GA61 does not), and four additional 
sites not contributing faunal data to my analysis but found in this area have 
associated radiocarbon dates (45FR39, 45FR47, 45WT36, 45VWV61). The frequency 
distribution of the 76 total dates, by 2000 year period, is shown in Fig. 2. Dates 
on charcoal are distinguished from those on freshwater mussel shell due to the 
debate concerning the validity of the latter (Chatters 1986; Sheppard et al. 1987). 
In this paper I presume shell dates are valid. Frequencies of dates per 2000 year 
period suggest stability of human occupational intensity until the last 2000 years, 
after which the intensity of occupation increased markedly. 

Frequencies of dated sites per 2000 year period suggest slightly lower occupa- 
tional intensity between 10,000 and 8000 B.P. and between 6000 and 4000 B.P. 
relative to the 10,000 to 8000 B.P. and 4000 to 2000 B.P. periods (Fig. 2). I suspect 
this results from the small sample of radiometrically dated sites for the early 
Holocene as there are three undated assemblages of artifacts temporally diagnostic 
of the early Holocene in this area (Brauner et al. 1990; Rice 1972). Thus there is 
no strong indication in the archaeological record for the Lower Snake River area 
of fluctuation in occupational intensity until the latest Holocene. Evidence of mid- 
Holocene change in occupational intensity may eventually be found in the Lower 
Snake as Chatters (1982) and Galm et al. (1981) present evidence indicating that 
the mid-Holocene absence of reliable water sources in the central, most arid part 
of the Columbia Basin resulted in human abandonment of that area during the 
Altithermal, and they suggest people moved to the canyons of major rivers at 
this time. 

The total NISP per taxon correlates with both the number of temporal incre- 



ments 



producing remains of a taxon (r s = 0.76, p 



0.66, V = 0.0005) and with 



^gesting any 

pparent faunal turnover is probably a function of sample size (fable 3). The 
number of breaks in temporal continuity is correlated inversely with the total NISP 
per taxon (r s = -0.42, p = 0.02) but perhaps not significantly with the number 
of sites producing remains of a taxon (r s - -0.30, p = 0.09). These latter two coef- 
ficients also suggest caution is warranted before concluding faunal turnover took 
place. For example, taxa such as the western harvest mouse (Reithrodontomys 
megalotis) and the western jumping mouse (Zapus princeps), both of which occur 
only during the latest Holocene, and pygmy rabbits, wolves (Canis latrans), river 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 47 



40 



C/3 



30 



C3 




O 20 
Z 10 







0.1-2 2-4 4-6 6-8 8-10 

Radiocarbon Yr B.P. (xlOOO) 



FIG. 2.— Frequencies of radiocarbon dates and dated sites in the Lower Snake 



River area. 



otters (Lutra canadensis), and raccoons (Proa/on lotor), which are reported only for 
the last 4000 years may not be represented in earlier time periods because 87.2% 
of the total Holocene NISP dates to the last 4000 years. 

Further evidence that sample size is creating indications of faunal turnover 
is found in the fact that adding the data in Table 4 to that in Table 3 results in 
the omission of mid- Holocene gaps in the occurrence of beaver (Castor canaden- 
sis), vole (Microtus sp.), Great Basin pocket mouse (Perognathus parvus), deer mouse 
(Peromyscus maniculatus), bobcat/lynx (Lynx sp.), marten (Maries americana) , badger 
(Taxidea taxus), and bighorn sheep. Other data indicate the red fox (Vulpes vulpes) 
also was present in eastern Washington throughout the Holocene (Lyman 1991b). 
While summing the two areal samples significantly increases the geographic area 
represented, it indicates Holocene faunal turnover in eastern Washington was 
probably minimal. Thus, instances of faunal turnover might be found in Table 3, 
but those are masked by the number of faunal specimens identified (NISP) and 
the number of sites sampled. The only clear evidence that faunal turnover took 
place involves the arctic fox (Alopex lagopus), which is recorded only for the early 
Holocene even though only 3.1% of the total Holocene NISP for the Lower Snake 
River area falls in that time period. 

Taxonomic richness per temporal period ranges from a high of 30 between 
2000 and 100 B.P. to a low of 15 between 6000 and 4000 B.P. (Table 3). In fact, 
taxonomic richness decreases steadily across the three temporal periods span- 
ning the early and mid-Holocene, increasing only after 4000 B.P. This temporal 
synchrony with first increased aridity and then with decreased aridity conforms 
with hypothesis (4). Richness is lowest between 8000 and 4000 B.P., coincident 
with the Altithermal period, precisely when hypothesis (3) suggests richness 



48 



LYMAN 



Vol. 12, No. 1 



should be lowest. Before those hypotheses are accepted, however, sample-size 
effects must be considered. 



TABLE 3.— Mammalian fauna for Lower Snake River. Values are number of 
identified specimens (NISP). 



Years B.P. 



10,000- 8000- 6000- 4000- 2000- 



Taxon 



8000 



6000 



4000 



2000 



100 



Nof 
NISP Sites 



SMALL MAMMALS NOT USED AS FOOD 



Spermophilus washingtoni 
(Washington ground squirrel) 

Spermophilus columbianus 
(Columbian ground squirrel) 

Thomomys talpoides 68 

(northern pocket gopher) 

Perognathus parvus 38 

(Great Basin pocket mouse) 

Reithrodontomys megalotis 
(Western harvest mouse) 

Peromyscus maniculatus 35 

(deer mouse) 

Neotoma cinerea 23 

(bushy-tailed woodrat) 

Microtus spp. 14 

(vole) 

Lagurus curtatus 
(sagebrush vole) 

Zapus princeps 
(western jumping mouse) 

CARNIVORES 

Cam's cf. latrans 
(coyote) 

Canus lupus 
(gray wolf) 

Alopex lagopus 6 

(arctic fox) 

Vulpes vulpes 5 

(red fox) 

Ursus atnericanus 
(black bear) 



4 



5 



17 



72 



55 



2 



3 



2 



4 



36 



15 



4 



32 



4 



3 



5 



94.5 



86 



216 



1.5 



1 



2.5 



12 



15.5 



37.5 



1 



0.5 



132.5 



10 



177 



190.5 



165 



5 



13.5 



17 



13.5 



2 



85.5 



2 



0.5 



1 



259 



172 



548 



230 



165 



43 



46 



45 



40 



2 



169 



3 



6 



6 



1 



8 



2 



9 



5 



1 



4 



4 



9 



5 



1 



10 



2 



1 



2 



1 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 49 



Years B.P. 



10,000- 8000- 6000- 4000- 2000- N of 
Taxon 8000 6000 4000 2000 100 NISP Sites 



Procyon lotor 
(raccoon) 

Martes americana 3 

(marten) 

Mustela frenata 4 

(long-tailed weasel) 

Taxidea taxus 
(badger) 

Lutra canadensis 
(river otter) 



11 2 2 



(bison) 

Ovis canadensis 
(mountain sheep) 



N of Sites 



14 3 7 8 



Richness 20 18 15 26 30 

Diversity 2.469 2.167 1.962 2.158 1.873 



3 1 



3 7 3 



1 3 12 16 4 



1.5 1.5 3 3 



Lynx sp. 1 1 5.5 5.5 13 6 

(bobcat/lynx) 



1 47 48 2 



MAMMALS USED AS FOOD 

Brachylagus idahoensis 
(pygmy cottontail) 

Sylvilagus nuttallii 2 23 13 34 258 330 8 

(Nuttall's cottontail) 

Lepus sp. 12 36 5 5 2,808 2,866 7 

(jackrabbit) 

Marmota cf. flaviventris 14 10 1 3.5 3.5 32 4 

(yellow-bellied marmot) 

Castor canadensis 2 1 22.5 3.5 29 6 

(beaver) 

Ondatra zibethicus 2 4 2 2 7 17 3 

(muskrat) 

Cervus elaphus 6 99 8 47.5 479.5 640 11 

(wapiti) 

Odocoileus spp. 29 226 57 284.5 937.5 1,534 11 

(deer) 

Antilocapra americana 16 84 65.5 112.5 1,247 1,525 10 

(pronghorn antelope) 

Bison bison 



0.5 367.5 28 3% 7 



4 5.5 39.5 49 6 



NISP 289 679 216 1,364.5 6,696.5 9,245 



50 



LYMAN 



Vol. 12, No. 1 



TABLE 4 . —Mammalian fauna for Upper-Middle Columbia River. Values are 
number of identified specimens (NISP). 



Years B.P. 



Taxon 



7000- 
4000 



1,201 



SMALL MAMMALS NOT USED AS FOOD 

Sorex sp. 
(shrew) 

Aplodontidae 
(mountain beaver) 

Eutamias sp. 
(chipmunk) 

Spermophilus sp. 12 

(ground squirrel) 

Glaucomys sp. 
(flying squirrel) 

Thomomys talpoides 
(northern pocket gopher) 

Perognathus parvus 
(Great Basin pocket mouse) 

Reithrodontomys megalotis 
(western harvest mouse) 

Peromyscus maniculatus 22 

(deer mouse) 

Neotoma cinerea 
(bushy-tailed woodrat) 

Microtus spp. 23 

(vole) 

Lagurus curtatus 21 

(sagebrush vole) 



341 



3 



4000- 
2000 



4 



2 



1 



85 



1 



1,576 



429 



1 



87 



9 



82 



73 



2000 
100 



22 



253 



198 



1 



25 



8 



35 



65 



Nof 
NISP Sites 



4 



2 



1 



119 



1 



3,031 



968 



2 



134 



20 



140 



159 



3 



1 



1 



13 



1 



24 



20 



1 



17 



7 



18 



15 



CARNIVORES 

Canis cf. latrans 
(coyote) 

Canis lupus 
(gray wolf) 

Vulpes vulpes 
(red fox) 

Ursus americanus 
(black bear) 

Ursus arctos 
(grizzly bear) 



44 



3 



8 



131 



5 



7 



9 



1 



45 



1 



4 



220 



9 



7 



21 



1 



18 



4 



3 



5 



1 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



51 



Taxon 



Years B.P. 



7000- 
4000 



4000- 
2000 



2000- 
100 



Nof 
NISP Sites 



Procyon lotor 
(raccoon) 

Martes americana 
(marten) 

Martes pennanti 
(fisher) 

Mustek frenata 
(long-tailed weasel) 

Tax idea tax us 
(badger) 

Mephitis mephitis 
(striped skunk) 

Lutra canadensis 
(river otter) 

Felis concolor 
(cougar) 

Lynx cf. rufus 
(bobcat) 



2 



55 



8 



5 



1 



6 



3 



4 



4 



2 



4 



3 



4 



4 



5 



7 



1 



1 



8 



7 



8 



64 



10 



7 



4 



9 



1 



3 



1 



4 



7 



2 



2 



1 



2 



MAMMALS USED AS FOOD 

Sylvilagus cf. nuttallii 
(Nuttall's cottontail) 

Lepus sp. 
(jackrabbit) 

Marmota cf . flaviventris 
(yellow-bellied marmot) 

Castor canadensis 
(beaver) 

Ondatra zibethicus 
(muskrat) 

Erethizon dorsatum 
(porcupine) 

Cervus elaphus 
(wapiti) 

Odocoileus spp. 



(d 




) 



Antilocapra americana 
(pronghorn antelope) 



11 



16 



397 



10 



17 



68 



47 



1,837 



95 



5 



30 



315 



44 



4 



8 



102 



4,692 



146 



17 



9 



194 



15 



2 



1 



95 



3,838 



490 



33 



55 



906 



69 



23 



77 



244 



1 0, 367 



731 



9 



13 



24 



12 



4 



4 



17 



26 



17 



52 



LYMAN Vol. 12, No. 1 



TABLE 4.— Mammalian fauna for Upper-Middle Columbia 
number of identified specimens (NISP). (continued) 



Years B.P. 



Taxon 



7000- 4000- 2000- N of 

4000 2000 100 NISP Sites 



Bison bison 
(bison) 

Ovis canadensis 
(mountain sheep) 



11 26 2 39 5 



411 1,103 560 2,074 24 



NISP 

N of Sites 

Richness 

Diversity 



4,669 9,005 5,901 19,575 

13 21 10 

25 36 27 

1.868 1.643 1.138 



Both taxonomic richness per temporal period and the number of sites pro- 
ducing faunal remains per temporal period in the Lower Snake River area are 
correlated with the total NISP per temporal period (for both, rs = 0.90, p = 0.07). 
These coefficients suggest richness may be a function of sample size and thus 
cannot be used in any straightforward manner to infer natural faunal change or 
change in human subsistence. As suggested earlier, if richness was lower during 
the mid-Holocene, human foragers may broaden their niche and take additional 
animal resources, a more diverse array of mammalian resources, more smaller 
mammalian resources, or some combination of these. Shannon diversity index 
values for the total mammalian fauna per temporal period for the Lower Snake 
and Upper-Middle Columbia faunas (Tables 3 and 4), when considered together, 
correlate with sample size (r s - -0.738; p - 0.05), indicating mid-Holocene diver- 
sities appear to be more even than early or late Holocene diversities simply because 
the mid-Holocene assemblages are small. Diversity of mammalian faunas, then, 
provides no trustworthy evidence for changes in mammalian faunas or human 
subsistence strategies. 

Prehistoric hunters probably did not, however, take all of the taxa listed in 
Tables 3 and 4, but it is difficult to determine from the published record precisely 
which taxa were regularly taken and which specimens listed in the tables we 
accumulated and deposited by human hunters. It seems likely that many 
specimens of the f ossorial taxa such as gophers and ground squirrels (Spermopni 
sp.) were naturally deposited, and perhaps some of the carnivore remains repre- 
sent individuals that died of natural causes on the sites. Marmots (Marmota flam- 
ventris), hares (Lepus sp.), rabbits (Brachylagus idahoensis and Sylvilagus nuttalw)^ 
beavers, and muskrats (Ondatra zibethicus) were clearly exploited by prehis o 
peoples in eastern Washington, as were ungulates, based on archaeolog 
evidence such as butchering marks. Considering only those non-carnivorous 
which I believe were exploited by people (see Table 3), relative abundances 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 53 



the small taxa, from earliest to latest temporal period, are: 11.1%, 10.8%, 10.2%, 

5.1%, and 46.7%. These proportions do not correlate with the NISP per temporal 

period (r s = 0.3; p >0.2) or with the number of sites producing faunal remains 

per temporal period (r s = 0.1, p>0.2), suggesting they are not a function of 

sample size. These relative abundances thus provide circumstantial evidence that 

prehistoric peoples did not shift to taking more smaller mammals during the mid- 

Holocene, but in fact decreased the frequency with which they were taken until 

the last 2000 years when, apparently, increased human populations (as indicated 

by measures of occupational intensity) resulted in increased competition and such 

small mammals became very important in the subsistence quest. 

The MSP of carnivore remains per temporal period is correlated with both 

the total NISP per time period (r s = 1.0, p = 0.05) and the number of sites 
producing mammal remains per time period (r$ = 0.90, p = 0.10), suggesting 
frequencies of carnivore remains are a function of sample size. (Note that in all 
cases the arctic fox is not included.) The diversity of carnivore taxa per time period 
is not correlated with the carnivore NISP per time period (r s - 0.20, p>0.20). 
Carnivore diversity is highest during the 10,000 to 8000 B.P. period (diversity = 
1.507), and lowest from 8000 to 6000 B.P. (diversity = 0.241). It then increases 
between 6000 and 4000 B.P. (diversity = 0.688), and is higher still in the late 
Holocene (4000 to 2000 B.P. diversity = 0.987; 2000 to 100 B.P. diversity = 1.099). 
Carnivore richness per period is not correlated with carnivore NISP per period 
or the number of sites producing mammal remains per period (r s = 0.825, p>0.1 
for both). The carnivore remains make up, from earliest to latest temporal period, 
6.2%, 5.6%, 3.0%, 3.7%, and 1.8% of the total mammal remains. I suspect that 
these statistics are reflecting (a) relatively low carnivore biomass during the mid- 
dle Holocene, and (b) decreased hunting of carnivores during the late Holocene, 
especially during the last 2000 years, as human foragers turned to other resources 
due to competition (see below). 

Relative frequencies of ungulates increase through the Holocene (until 2000 
B.P.) when only taxa utilized as food are considered; from earliest to latest these 
are: 88.9%, 89.2%, 89.8%, 94.9%, and 53.3%. This does not necessarily indicate 
that hypothesis (2) is wrong, but only that humans were not responding to 
hypothesized changes in the biomass of large mammals by taking such mammals 
in proportion to their hypothesized lower natural abundance. The drop in relative 
abundance of ungulates between 2000 and 100 B.P. corresponds to the increase 
in small mammal resources at that time. Again, this could well reflect the increased 
competition for resources that may have occurred as a result of increased inten- 
sity of human occupation at this time (refer to Fig. 2). 

Ungulate NISP per temporal increment is correlated with total NISP of all 
mammals per temporal increment (r s - 0.90, p = 0.07), suggesting ungulate NISP 
may be a function of sample size. From earliest to latest temporal penod, ungulate 
remains make up 17.6%, 60.8%, 60.6%, 59.9%, and 40.8% of the total mammalian 
NISP. These proportions are precisely the opposite of that indicated in hypothesis 
(2); they do not correlate with total NISP (r s = 0.2; p>0.2) and thus do not seem 
to be a function of sample size. Apparently Gustafson (1972) was correct in noting 
that relatively more ungulate remains were deposited during the mid-Holocene 
than prior or subsequent to that time, but contrary to his belief those higher 



54 



LYMAN Vol. 12, No. 1 



frequencies do not appear to be attributable to increased human occupational 
intensity. I can offer no empirically warranted explanation for the changing fre- 
quencies of ungulate remains relative to the total mammalian fauna, but wonder 

if they might be attributable to a shift in human subsistence practices resulting 
in small mammals being gradually replaced by plants and fish during the mid- 

Holocene, which would result in fewer small mammal remains being deposited, 
especially during the mid-Holocene. There are, however, too few botanical and 
fish data to assess this possibility directly. Archaeological evidence indicates there 
was an increase in the relative abundance of plant processing and fishing tools 
during the mid-Holocene (e.g., Bense 1972; Galm et al. 1981) which could be taken 
as circumstantial evidence for low availability of mammalian biomass, especially 
small taxa (and carnivores?), during the mid-Holocene. After 4000 B.P., people 
continued to increase their exploitation of plants and fish, largely omitting small 
mammals and focusing more on large mammals, particularly ungulates. 

Shannon diversity index values for only those mammals clearly exploited as 
food resources are, from earliest to latest: 1.725, 1.478, 1.349, 1.471, and 1.362. 
These do not correlate with NISP per temporal period (r s = -0.4, p>0.2), and 
thus seem to indicate decreasing diversity and increasing specialization on 
mammals between 10,000 and 4000 B.P., with only slightly more diversity 
between 4000 and 2000 B.P. after which diversity decreases again, perhaps, as 
suggested, due to increasing utilization of fish and plant resources. 



Upper-Middle Columbia River.— Miss (1985:274) suggests that rates of accumu- 
lation of bone and lithic artifacts (measured as density of each per 1000 year 
temporal period) are "consistent after 4000 yr B.P/' in the aggregated Chief Joseph 
Dam Reservoir sites. Prior to that time accumulation of artifacts was slower. If 
those accumulation rates are indicative of occupational intensity, then such 
intensity was lower prior to 4000 B.P. than after that time. Similar analyses of 
the Wells Reservoir materials cannot be performed, but the total radiocarbon dates 
for the Upper-Middle Columbia River area seem to conform to Miss's (1985) in- 
terpretation.. As with the lower Snake River radiocarbon record, I compiled all 
radiocarbon dates for sites in the Upper-Middle Columbia River area known to 
me regardless of whether or not those dates were from sites producing fauna! 
remains for this study. That resulted in 229 dates from 53 sites; 25 of the 28 sites 
contributing faunal data to my analysis have radiocarbon dates. As shown in 
Fig. 3, the frequency of radiocarbon dates is lowest between 7000 and 4000 B.P., 
but those frequencies are essentially equal for the 4000 to 2000 B.P. and 2000 to 
100 B.P. time periods. As well, the frequencies of dated sites per temporal period 
suggest fewer sites were occupied prior to 4000 B.P. than after that time, and 
that the number of occupied sites was essentially the same for the latest two 
temporal periods. Thus I take occupational intensity to have been somewhat lower 
between 7000 and 4000 B.P. than subsequent to 4000 B.P. 

The total NISP for this area is larger than that for the Lower Snake River area, 
but the Upper-Middle Columbia River faunal record spans only the last 70 k o 
the Holocene. Presuming this means that the Upper-Middle Columbia River 
sample is more representative of the mid- and late Holocene than the Lower Sna 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 



55 



sample is, there should be less evidence of sampl 
?mblages. MSP per taxon is, however, correlated v 



remains 



temporal 



<0 



number of sites producing remains 



v<0 



number of temporal 



0.81, /X0.0001). 



These correlations suggest the appearance of faunal turnover in Table 4 mav be 



sam 



otter between 2000 and 100 B.P. because only 30% of the total NISP dates to that 
time period and the 4000 to 2000 B.P. sample is almost 53% larger than the 2000 
to 100 B.P. samole. 



100 



80 



C/5 



Q 60 

40 




a 



20 







0.1-2 2-4 4-7 

Radiocarbon Yr B.P. (xlOOO) 



FIG. 3.— Frequencies of radiocarbon dates and dated sites in the Upper-Middle 
Columbia River area. See legend for Figure 2. 

Three taxa represented in the two latest samples are not represented in the 
7000 to 4000 B.P. sample: western harvest mouse, fisher (Martes pennanti), and 
long-tailed weasel (Mustek frenata) . As well, two taxa represented in the earliest 
samples are not represented between 4000 and 100 B.P.: striped skunk (Mephitis 
mephitis) and marten. Five of the six taxa (including river otters) mentioned thus 
far are mustelids, and it is perhaps significant, then, that one of the Wells Reser- 
voir sites not included in my analysis or in Table 4 due to small sample size (NISP 
<30) produced a single specimen of wolverine (Gulo luscus) dating between 7000 
and 4000 B.P. (Chatters 1986). The significance of this is probably not that six 
of these seven taxa are mustelids, but rather that they are carnivores, taxa which 
are typically rare in eastern Washington archaeological sites. For example, four 
of the eight taxa represented only in the 4000 to 2000 B.P. assemblage are carni- 
vores: cougar (Felis concolor), raccoon, grizzly bear (Ursus arctos), and red fox. That 



56 



LYMAN Vol. 12, No. 1 



is, of the 14 taxa not continuously represented between 7000 and 100 B.P., nine 
(64%) are carnivores (10 of 15, or 67%, if the wolverine is included). Thus if faunal 
turnover took place along the Upper-Middle Columbia River during the last 7000 
years, I do not believe it is clearly reflected in Table 4. What appears to be faunal 
turnover in this table is more likely a reflection of sample size effects. 

Carnivore NISP per temporal period does not reflect the total NISP per 
temporal period (I did not calculate correlation coefficients because there are 
only three cases), but carnivore NISP per period does reflect the number of sites 
sampled per period. Neither carnivore richness per period nor carnivore diver- 
sity per period mirrors carnivore NISP per period. Carnivore diversity is greatest 
between 7000 and 4000 B.P. (diversity = 1.366), and progressively decreases 
through time (diversity between 4000 and 2000 B.P. = 1.223; diversity between 
2000 and 100 B.P. = 1.101). As in the Lower Snake River area samples, carnivores 
decrease in abundance from the middle Holocene (7000 to 4000 B.P.) when they 
account for 2.7% of all mammalian remains, to 2.0% between 4000 and 2000 B.P. 
and 1.2% between 2000 and 100 B.P. Faunal turnover and changes in biomass 
are not apparent here, but the decreased abundance of carnivores may reflect 
changes in human subsistence (see below). 

While I did not calculate correlation coefficients here because there are only 
three cases, it is important to note that taxonomic richness per temporal period 
is perfectly correlated with total NISP per temporal increment (Table 4). This 
suggests richness per temporal period may be an effect of sample size. Recall that 
optimal foraging theory suggests dietary breadth (richness and/or diversity) will 
increase as resources become scarce or as competition for resources increases. 
Thus, during the mid-Holocene along the Upper-Middle Columbia, if resources 
were scarce but occupational intensity was relatively low, dietary breadth at that 
time may have been similar to dietary breadth during the late Holocene (last 4000 
years) when resources were more abundant but occupational intensity was 
relatively high. But the diversity of mammalian food resources decreases from 
the mid-Holocene to the latest Holocene; Shannon diversity index values for 
only those taxa used as food (same taxa as for Lower Snake), from earliest to latest 
are: 1.084, 0.925, 0.935. That could be taken as evidence that mammalian biomass 
was low during the mid-Holocene, and people broadened their niche by taking 
a more diverse set of mammalian resources during this time, but those index 
values are perfectly inversely correlated with the NISP of taxa used as food per 
temporal period, indicating that the smaller the NISP the greater the diversity 
index value. 

The proportions of mammals used as food that are small taxa are from earliest 
to latest 9.7%, 4.4%, and 4.0%. These relative frequencies do not correlate with 
the total NISP per temporal period, suggesting they are not a function of sample 
size. They indicate more small mammals were used as food between 7000 and 
4000 B.P. than subsequently, and this might be taken as circumstantial evidence 
for a mid-Holocene decrease in naturally available mammalian biomass. These 
proportions do not seem to be a result of intensity of human occupation as 
occupational intensity was apparently low during the mid-Holocene. And, as witn 
the Lower Snake River sample, when only mammalian taxa used as food are con- 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 57 



sidered, the relative abundance of ungulates increases from earliest to latest sue- 



more specialized economy 



terms of mammalian 



It was suggested that in the Lower Snake River area, fish and plants became 

£» imnnrfanf rncAiir^nr T..i*-U 4-U^. ^_~ _r A It'll t «« . • . 



more important resources with the emergence 



some 1m 






remains identified from sites in the Upper-Middle Columbia River area, 32% date 
between 7000 and 4000 B.P., 57.3% date between 4000 and 2000 B.P., and 10.7% 
date between 2000 and 100 B.P. (fish data compiled like mammalian data, but 
without regard for specific site sample sizes). Presuming fish were used as food 
by prehistoric human occupants of the area (fish were important ethnographi- 
cally), from earliest to latest fish make up 28%, 47.9%, and 38.8% of the summed 
food mammal and fish remains. These proportions suggest fish were a more 
important food resource after 4000 B.P. than during the mid-Holocene, but their 
importance during the early Holocene in this area is unknown. Thus it seems 
small mammals decreased in importance while large mammals, especially 



after the mid-Holocene 



importance in the Uooer-Middle 



remains 



mammalian 



complete faunas are, from earliest to latest, 51.4%, 67.4%, and 84.5^ 
do not correlate with the total mammalian NISP. These values 
hypothesis (2), that ungulate biomass was at its lowest during the mi 
Why these frequencies increase through time probably has to do i 
tition. The low occupational intensity and perhaps low mammal 



half 



deposited mammalian remains 



intensity increased, competition increased, and ungulate biomass increased as 
well. The relative frequencies of ungulate remains suggest the latter may have 
out-paced the former two variables, permitting an increasingly specialized 
mammalian subsistpnrp hasp rnnsistint* lamplv of uneulates. 



CONCLUSIONS 



The possible responses of mammalian communities to changing Holocene 
climates and of human hunter-gatherers to fluctuating resource abundances led 
researchers working 30 years ago in eastern Washington to suggest that (a) mam- 
malian biomass was at its lowest level during a mid-Holocene period of relative 
aridity and (b) that humans would respond by taking a broader range of resources. 
Those suggestions are easily phrased as hypotheses, but in order to be tested 
they must take into account the effects of sample size on measures of taxonomic 
richness, diversity, and turnover, and the effects of human occupational inten- 
S1 ty on measures of animal abundances. 

Frequencies of radiocarbon dates and of dated sites indicate human occupa- 
tional intensity of the Lower Snake River area was relatively stable from 10,000 
B -P. until about 2000 B.P. Faunal turnover during the Holocene there is largely 



58 



LYMAN 



Vol. 12, No. 1 



obscured by sample-size effects but minimally entailed loss of the arctic fox in 
the early Holocene. Mammalian biomass appears to have been at its lowest level 
between 6000 and 4000 B.P., having decreased steadily since the end of the 
Pleistocene, much as was suggested 30 years ago, but problems of sample size 
make such a conclusion tenuous. Human occupational intensity does not account 
for a perceived steady increase in the relative abundance of ungulate remains 
between 10,000 and 2000 B.P.; rather, that increase may be a function of decreased 
deposition of the remains of non-ungulate mammals, especially those small 
mammals such as leporids and marmots that were humanly exploited. Gradually 
and throughout the Holocene human foragers took fewer and fewer small mam- 
mals. Archaeological data indicate the gradually increasing focus on ungulates 
seems to have accompanied a shift in subsistence pursuits to increased reliance 
on plant and fish resources in the last 4000 years. 

Rates of artifact deposition and frequencies of radiocarbon dates and dated 
sites indicate the Upper-Middle Columbia River area was less intensively occupied 
between 7000 and 4000 B.P. than after that time. Indications of faunal turnover 
there involve mostly carnivores, which are rare naturally and archaeologically, 
and thus seem to provide poor indications of change in the taxonomic compo- 
sition of the mammalian community. Stability in abundances of faunal remains 
relative to site frequencies and taxonomic richness can be accounted for by a 
combination of low occupational intensity and low mammal biomass during the 
mid-Holocene, and by a combination of higher occupational intensity and higher 
mammalian biomass during the last 4000 years. As with the Lower Snake River 
fauna, human foragers living along the Upper-Middle Columbia River seem to 
have slowly shifted their mid-Holocene subsistence pursuits to taking more 
ungulates and fish in the late Holocene. 



ACKNOWLEDGEMENTS 



I thank J.R. Galm for helping me keep up with archaeological research in eastern 
Washington, Cesar Veintimilla Bustamante for preparing the Spanish abstract, and Brigitte 
Holt for preparing the French abstract. An early draft received editorial help from M.J. 
O'Brien. Comments on later drafts by R.W. Graham, D.K. Grayson, J.I. Mead, and T.R. 
van Devender helped me polish the rough spots. 



LITERATURE CITED 



BADGLEY, CATHERINE AND PHILIP D. 
GINGERICH. 1988. Sampling and faunal 
turnover in early Eocene mammals. 
Palaeogeography, Palaeoclimatology, 
Palaeoecology 63:141-157. 

BARNOSKY, CATHY W., PATRICIA M. 
ANDERSON AND PATRICK J. BART- 
LEIN. 1987. The northwestern U.S. dur- 
ing deglaciation: Vegetational history 
and paleoclimatic implications. Pp. 289- 
321 in The Geology of North America, 



Vol. K-3: North America and Adjacent 
Oceans During the Last Deglaciation. 
W.F. Ruddiman and H.E. Wright, Jr. 
(editors). Geological Society of Amenca, 

Boulder. 
BAUMHOFF, MARTIN A. AND ROBERT F. 
HEIZER. 1965. Postglacial climate and 
archaeology in the Desert West. Pp 697- 
707 in The Quaternary of the United 
States. H.E. Wright, Jr. and David G. 
Frey (editors). Princeton University 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



59 



LITERATURE CITED (continued) 



Press, Princeton. 



MIT 



age use, water consumption, and pro- 
ductivity of pronghorn antelope on 



Wildlife 



34 



BELOVSKY, GARY E. 1987. Hunter-gatherer 
foraging: A linear programming ap- 
proach. Journal of Anthropological 
Archaeology 6:29-76. 

1988. An optimal foraging- 

based model of hunter-gatherer popula- 
tion dynamics. Journal of Anthropolog- 
ical Archaeology 7:329-372. 
BENSE, JUDITH A. 1972. The Cascade 
Phase: A Study in the Effect of the Alti- 
thermal on a Cultural System. Unpub- 
lished Ph.D. dissertation, Department of 
Anthropology, Washington State Univer- 
sity, Pullman. 

BETTINGER, ROBERT L. 1991. Hunter- 
Gatherers: Archaeological and Evolu- 
tionary Theory. Plenum Press, New 
York. 

BRAUNER, DAVID, R. LEE LYMAN, HAL 
GARD, STEPHAN MATZ AND REBEC- 



McCLELLAND. 1990 



45WT134 



Whitman County, Washingt 
to the U.S. Army Corps of Engineers, 
Walla Walla District. Oregon State Uni- 
versity, Department of Anthropology, 
Corvallis. 



Examination 



some faunal remains from the Marmes 

Rockshelter Floodplain. Unpublished 
Masters thesis, Department of Anthro- 



Pullman. 



Washingt 



CHATTERS, JAMES C. 1982. Prehistoric 
settlement and land use in the dry 
Columbia Basin. Northwest Anthro- 
pological Research Notes 16:125-147. 
1984a. Dimensions of site 



structure: The 



two sites in Okanogan County, Washing- 
ton. Report to the U.S. Army Corps of 
Engineers, Seattle District. Central Wash- 
ington Archaeological Survey, Ellens- 
burg. 



1984b. Human adaptation 

along the Columbia River, 4700-1600 



B.P.: A report of test excavation at River 

Mile 590, North Central Washington. 
Central Washington University, Research 
Reports 84-1. Ellensburg. 

1986. The Wells Reservoir 

Archaeological Project, Washington, 
Vol. 1: Summary of findings. Central 
Washington Archaeological Survey, 
Archaeological Report No. 86-6. Ellens- 
burg. 

COLSON, ELIZABETH. 1979. In good years 
and in bad: Food strategies of self-reliant 
societies. Journal of Anthropological 
Research 35:18-29. 

DAUBENMIRE, R. 1970. Steppe vegetation 
of Washington. Washington Agricultural 
Experiment Station, Technical Bulletin 

62. Pullman. 
EARLE, TIMOTHY K. 1980. A model of sub- 
sistence change. Pp. 1-29 in Modeling 
Change in Prehistoric Subsistence Strate- 
gies. Timothy K. Earle and Andrew L. 
Christenson (editors). Academic Press, 

New York. 
ESTES, RICHARD D. 1979. Ecological 
aspects of bighorn sheep populations 
in southeastern Washington. Unpub- 
lished Masters thesis, Department of 
Zoology, Washington State University, 

Pullman. 
FRANKLIN, JERRY F. AND C.T. DYRNESS. 
1973. Natural vegetation of Oregon and 
Wachmoinn TISRA. Forest Service, 



PNW 



land. 



AND RICHARD 



DAUGHERTY. 1963. Late 



and archaeological 
chronology'of the Columbia Plateau, 
Washington. Washington State Univer- 
sity, Laboratory of Anthropology, 
Reports of Investigations No. 23. 

Pullman. 

LM, JERRY R., GLENN D. HART- 
MANN, RUTH A. MASTEN AND 
GARRY O. STEPHENSON. 1981. A 
cultural resources overview of Bonne- 
ville Power Administration's Mid- 



Wash 



Washingt 



100 



Cheney. 



60 



LYMAN 



Vol. 12, No. 1 



LITERATURE CITED (continued) 



GRAYSON, DONALD K. 1984. Quantita- 
tive Zooarchaeology. Academic Press, 
New York. 

GUSTAFSON, CARL E. 1972. Faunal 

remains from the Marmes Rockshelter 
and related archaeological sites in the 
Columbia Basin. Unpublished Ph.D. 
dissertation, Department of Zoology, 
Washington State University, Pullman. 

HAYDEN, BRIAN. 1981. Subsistence and 
ecological adaptations of modern hunter/ 
gatherers. Pp. 344-421 in Omnivorous 
Primates. Robert S.O. Harding and Geza 
Teleki (editors). Columbia University 
Press, New York. 

JOHNSON, ROLF L. 1983. Mountain goats 
and mountain sheep of Washington. 
Washington State Game Department, 
Biological Bulletin No. 18. Olympia. 

KEENE, ARTHUR S. 1985. Nutrition and 
economy: Models for the study of pre- 
historic diet. Pp. 155-190 in The Analysis 
of Prehistoric Diets. Robert I. Gilbert, 
Jr. and James H. Mielke (editors). Aca- 
demic Press, Orlando. 

KENASTON, MONTE RAY. 1966. The 
archaeology of the Harder Site, Franklin 
County, Washington. Washington State 
University, Laboratory of Anthropology, 
Report of Investigations No. 35. Pullman. 

KITCHEN, DAVID W. AND BART W. 
O'GARA. 1982. Pronghorn Antilocapra 
americana. Pp. 960-971 in Wild Mammals 
of North America. Joseph A. Chapman 
and George A. Feldhamer (editors). 
Johns Hopkins University Press, Balti- 
more. 

KOCH, CARL F. 1987. Prediction of sample 
size effects on the measured temporal 
and geographic distribution patterns of 
species. Paleobiology 13:100-107^ 

LIVINGSTON, STEPHANIE D. 1985. Sum- 
mary of faunal data. Pp. 365-419 in 
Summary of Results, Chief Joseph Dam 
Cultural Resources Project, Washington. 
Sarah K. Campbell (editor). Report to the 
U.S. Army Corps of Engineers, Seattle 
District. University of Washington, Office 
of Public Archaeology, Seattle. 

LYMAN, R. LEE. 1986a. On the analysis 
and interpretation of species list data 
in zooarchaeology. Journal of Eth no- 
biology 6:67-81. 



. 1986b. On the Holocene 

history of Ursus in eastern Washingt 
Northwest Science 60:67-72. 

eology 



1987. 



taphonomy: A general consideration. 
Journal of Ethnobiology 7:93-117. 

1988. Zooarchaeology of 



45D0189. Pp. 97-141 in Archaeological 
Investigations at River Mile 590: The 
Excavations at 45D0189. Jerry R. Galm 
and R. Lee Lyman (editors). Eastern 
Washington University, Reports in 



burg. 



dogy and History 100 



1990. Zooarchaeology. Pp. 



98-138 in Archaeology Data Recovery 
at Hatiuhpuh, 45WT134, Whitman 
County, Washington. David Brauner 
(editor). Report to the U.S. Army Corps 
of Engineers, Walla Walla District. Ore- 
gon State University, Department of 

Anthropology, Corvallis. 

. 1991a. Late Quaternary bio- 



geography of the pygmy rabbit (Brachy- 
lagus idahoensis) in eastern Washington. 
Journal of Mammology 72:110-117. 



• • 



The 



of the red fox (Vulpes vulpes) in eastern 



Wash 
26. 



STEPHANIE 



INGSTON. 1983. Late Quaternary mam- 
malian zoogeography of eastern Wash- 
ington. Quaternary Research 20:360-373. 

McCORQUODALE, SCOTT M. 1985. Archae 
ological evidence of elk in the Columbia 
Basin. Northwest Science 59:192-197. 

__, KENNETH J. RAEDEKE AND 

RICHARD D. TABER. 1986. Elk habitat 
use patterns in the shrub-steppe ot 
Washington. Journal of Wildlife Manage 
ment 50:664-669. 

MEHRINGER, PETER J., JR. 1985. Late- 
Quaternary pollen records from the 
interior Pacific Northwest and northern 
Great Basin of the United States. Pp. 
167-189 in Pollen Records of Late- 
Quaternary North American Sediments. 
Vaughan M. Bryant, Jr. and Richard U 
Holloway (editors). American Associa- 
tion of Stratigraphic Palynologists, 

Dallas. 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



61 



LITERATURE CITED (continued) 



MILLER, STANLEE M. 1977. Mammalian 
remains from the Juniper Forest Pre- 
serve, Franklin County, Washington. 
Unpublished Masters thesis, Department 
of Zoology, University of Idaho, 
Moscow. 

MISS, CHRISTIANS 1985. Site frequency, 
intensity of use, and differentiation 
through time. Pp. 269-286 in Summary 
of Results, Chief Joseph Dam Cultural 
Resources Project, Washington. Sarah K 
Campbell (editor). Report to the U.S. 
Army Corps of Engineers, Seattle Dis- 
trict. University of Washington, Office 
of Public Archaeology, Seattle. 

NELSON, CHARLES, WARREN DeBORE 
AND DIANE GIFFORD. 1968. Faunal 
remains from Assemblage 3A, site 
45C01 . Manuscript on file, Department 
of Anthropology, Washington State 
University, Pullman. 

O'CONNELL, JAMES F. AND KRISTEN 
HAWKES. 1981. Alyawara plant use and 
optimal foraging theory. Pp. 99-125 in 
Hunter-Gatherer Foraging Strategies. 
Bruce Winterhalder and Eric Alden 
Smith (editors). University of Chicago 
Press, Chicago. 

OLSON, DEBORAH L. 1983. A descrip- 
tive analysis of the faunal remains from 
the Miller Site, Franklin County, Wash- 
ington. Unpublished Masters thesis, 
Department of Anthropology, Washing- 
ton State University, Pullman. 

PETERSON, CHARLES H. 1977. The paleo- 
ecological significance of undetected 
short-term variability. Journal of Paleon- 
tology 51:976-981. 

PIANKA, ERIC R. 1978. Evolutionary 

Ecology, 2nd ed. Harper & Row, New 
York. 

PITT, M.D. AND B.M. 



B.M. WIKEEM. 1978. 
Diet preference of California bighorn 
sheep on native rangeland in south- 
central British Columbia. Proceedings of 
the 1978 Northern Wild Sheep and Goat 
Conference, pp. 331-341. 

PULLIAM, H. RONALD. 1981. On predict- 

Journal of Ethnobiology 



1:61-68. 



Windust 



of Anthropology, Report of Investiga- 
tions No. 50. Pullman. 

RICK, JOHN W. 1987. Dates as data: An 
examination of the Peruvian Preceramic 
radiocarbon record. American Antiquity 
52:55-73. 

SCHALK, RANDALL F. 1983. General sum- 
mary and implications. Pp. 117-163 in 
The 1978 and 1979 Excavations at Straw- 
berry Island in the McNary Reservoir. 
Randall F. Schalk (editor). Washington 
State University, Laboratory of Archae- 
ology and History, Project Report 19. 
Pullman. 

AND DEBORAH L. OLSON. 

1983. The faunal assemblages. Pp. 75- 
110 in The 1978 and 1979 Excavations 
at Strawberry Island in the McNary 
Reservoir. Randall F. Schalk (editor). 
Washington State University, Laboratory 
of Archaeology and History, Project 
Report 19. Pullman. 

SCHINDEL, DAVID E. 1980. Microstrati- 
graphic sampling and the limits of pale- 
ontologic resolution. Paleobiology 6:408- 

426. 
SHEPPARD, JOHN C, PETER E. WIGAND, 
CARL E. GUSTAFSON AND MEYER 
RUBIN. 1987. A reevaluation of the 
Marmes Rockshelter radiocarbon chron- 
ology. American Antiquity 52:118-125. 

SIMMS, STEVEN R. 1987. Behavioral ecol- 
ogy and hunter-gatherer foraging: An 
example from the Great Basin. British 
Archaeological Reports International 
Series No 381. Oxford. 

VAN VUREN, DIRK. 1984. Summer diets of 
bison and cattle in southern Utah. Jour- 
nal of Range Management 37:260-261. 

. 1987. Bison west of the Rocky 



Mountains: An alternative explan 
Northwest Science 61:65-69. 

AND MARTIN P. BRAY 



Diets of bison and cattle on a seeded 
range in southern Utah. Journal of Range 
Management 36:499-500. 

1985. The recent geographic 



distribution of Bison bison in Oregon. 

relet 66:56-58. 
. 1986. Population dynamics of 



The 



in lower Snake River region prehistory. 
Washington State University, Laboratory 



bison in the Henry Mountains, Utah. 
Journal of Mammalogy 



62 



Vol. 12, No. 1 



LITERATURE CITED (continued) 



WINTERHALDER, BRUCE. 1986. Optimal 
foraging: Simulation studies of diet 
choice in a stochastic environment. 
Journal of Ethnobiology 6:205-223. 

YENT, MARTHA E. 1976. The cultural 



sequence at Wawawai (45WT39), Lower 
Snake River Region, Southeastern Wash- 
ington. Unpublished Masters thesis, 
Department of Anthropology, Washing- 
ton State University, Pullman. 



BOOK REVIEW 



The Mixe of Oaxaca. Religion, Ritual, and Healing. Frank J 
Texas: University of Texas Press, 1991. Pp. xx, 253. $35. 0( 
ISBN 0-292-76517-7. 



The product of extensive field work spanning five trips (one of which lasted 
16 months) and thorough bibliographic research, this book is indeed a welcome 
contribution to ethnobotany and anthropology of Oaxaca and particularly of one 
of the most poorly known native cultures of Mexico. It is not frequent that an 
Indian group of 76,000 individuals living in some 50 villages in a country where 
so much anthropological work has been carried on for so long has escaped such 
dedicated study as that apparent in this volume. As Dr. Munro Edmonson, 
anthropologist at Tulane University, states: "Mixe culture and Lippo's descrip- 
tion of it stands alone in the Middle American ethnography .... This work fills 
a gap in the literature for which there is no alternative and no competition." 

A long introduction acquaints the reader to the methodology used in the field 
studies. Then follows nine chapters: Social Organization and Kinship; Subsistence 
Agriculture; Religious Belief System; Calendrical System; Ritual Behaviour; Rites 
of Passage; Village Festivals; Medical Concepts and Behaviour; Postscript. There 
are three appendices: Mixe Region; Mixe Phonemes; Mixe Texts. Fifteen pages 
are devoted to two lists of names: one of Mixe terms, the other Spanish and 
Nahuatl words. The bibliography of literature cited enumerates 306 items. There 
is a detailed index. 

This masterly book is beautifully published and conservatively priced. 



Richard Evans Schultes 

Director Emeritus 

Botanical Museum of Harvard University 

Cambridge, Massachusetts 



/. EthnobioL 12(1) :63-80 



Summer 1*^92 



AN OPTIMAL FORAGING ANALYSIS 
OF PREHISTORIC SHELLFISH COLLECTING 
ON SAN CLEMENTE ISLAND, CALIFORNIA 



L. MARK RAAB 

Depa rtmen t of Anth ropology 

California State University 

Northridge, CA 91330 



ABSTRACT.— The subsistence yield of black abalones (Haliotis cracherodii) and 
black turban snails (Tegula funebralis) was estimated within prehistoric aboriginal 
shell middens dated 250-2830 B.P. on San Clemente Island, California. Abalones 
were the key element of the aboriginal shellfish economy, but consumption of 
the smaller turban snails increased with depletion of abalones in a pattern that 
conforms to an optimal foraging model of predation. 



RESUMEN.— El rendimiento de abalon negro (Haliotis cracherodii) y caracol de 
turbante negro (Tegula funebralis) para la subsistencia de las poblaciones indigenas 
fue estimado en concheros prehistoricos (250 a 2830 ahos antes del presente, 
datados con carbono radioactivo) en la Isla de San Clemente, California. Los 
abalones fueron el elemento clave en la economi'a aborigen de mariscos, pero el 
consumo de los caracoles de turbante, mas pequehos, aumento con el agotamiento 
de los abalones en un patron que concuerda con un modelo de depredacion como 
forrajeo optimo. 

RESUME.— Des traces de presence d'haliotides noires (Haliotis cracherodii) et 
d'escargots noirs a turban (Tegula funebralis) ont he decouvertes au sein de vestiges 
re'siduels prehistoriques de coquillages aborigines (250 a 2830 annees radio- 
carbones avant le temps pre'sent) sur lite de San Clemente, en California. Les 
haliotides constituaient l'element principal de I'economie de coquillages 
aborigenes, mais la consommation des escargots a turban, bien plus petits, s'est 
acme au fur et a mesure de la disparition des haliotides selon une progression 
conforme a un modele optimal de ravages causes par des predateurs. 



Marine gastropod species were important subsistence resources among the 
aboriginal peoples of the southern California Channel Islands. Two species, the 
black abalone (Haliotis cracherodii Leach, 1814) and the black turban snail (Tegula 
funebralis A. Adams, 1855), are the focus of the present discussion. Abalone shells 
are a conspicuous component of prehistoric middens (domestic refuse deposits) 
on the Channel Islands. Beads, ornaments, fishhooks, containers, and other 
artifacts were manufactured from abalone shells. Despite this presence, recon- 
struction of the snhcistPnrP role of abalones and other shellfish species remains 



develop 



Reconstructions usually 



i 



food yields based upon shell weight or MNI (minimum numbt 
Is) figures. Frequently, such estimates do not reflect the possibility 
secies may have been utilized in shifting patterns of exploitation 



64 



RAAB Vol. 12, No. 1 



time interval represented by a midden. This situation is surprising 



maritime 



undergoing reexamination in many quarters. Recent archaeological and 

ethnographic studies have demonstrated the important role that shellfishing may 

play within a variety of economic adaptations. Consideration of biological 

characteristics of shellfish species, combined with appropriate analytical methods, 

reveal dynamic patterns of aboriginal shellfish collecting on San Clemente Island, 
southern California. 1 



ABALONE AND TEGULA BIOLOGY 



Abalones are large, herbivorous marine snails that inhabit many regions of the 
world. They require rock surfaces, where they attach themselves with a large 
muscular "foot." Thus attached, they are protected by a thick univalve shell fron 
predators and other hazards while grazing on floating kelp fragments. Specie; 
are readily differentiated based upon shell morphology. Water drawn througl 
the gills is expelled through a series of prominent respiratory pores arrayed alonj 
the shell, the number and characteristics of which also vary by species (Howortl 
1988:38-44). Abalone species occupv much of the Pacific Coast of North America 



elsewhere 



(Morris 1966:52), as well as Australia, New 



Four species occur in significant frequencies within archaeological deposits 
of southern California. The three largest of these species generally occupy the 
subtidal zone; i.e., rocky substrates that remain submerged even during the lowest 
tides. Although these species vary in size and specific habitat requirements, they 
all inhabit substantially similar environments. Among the subtidal forms is the 
largest species in the world, the red abalone (Haliotis rufescens) with a shell length 
approaching 300 mm and a soft-tissue weight of as much as 3 lbs (1.36 kg; Morris 
et. al. 1980:232; Ault 1985:4). The green (H. fulgens) and pinkfH. corrugata) abalones 
reach a maximum length of about 250 mm (Ault 1985:4; Morris et. al. 



inhabit a depth gradient from 



m for the pink 



m tor the red abalone. 18 m 



ception in California, red abalone and other species may 



m 



most frequently between about 6 m and 24 m (Cox 1960:386-390; Ault 
5-16). A number of factors appear to affect preferred depth, including algal 



offers protection 



et al. 1980:232). 



water temperature (Ault 1985:15-16; Morris 



subtidal 



Where 



species exist in comparatively shallow water, diving can produce large harvests. 
The relatively large size of the subtidal forms may compensate for the effort 
involved in diving. It has also been suggested that red abalone may have been 
intertidal during the early and mid-Holocene (i.e., around 5000-7000 B.P.; Glassow 
et al. 1988:70), when periods of sea temperature cooler than at present may 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 65 



have existed in the California Bight (southeastward bend of the California coast 



grip on its rocky perch. 



prying device (aboriginal forms w 
i to break the animal's extremely 



Water temperature has a major im 



The 



exampl 



species 



Water temperature requirements 



more im 



assessment of dietary yield. Use of abalone species as indicators of past 



environmental regimes is a topic im 



(Walker and Snethkamp 1984 



1977 



sperm are 
, larvae be 



mm 



-tin the first year (Morris et. al 1980:232-237; Ault 1985:5-6). After reaching 
juvenile stage, abalone grow more slowly, adding a few millimeters length 
i year. Abalone appear to reach sexual maturity within 3-5 years, and it is 
light that few individuals live longer than 20 years (Morris et al. 1980:232-237). 
abalone species are slow growing organisms with few defenses against human 
dators. Consequently, their numbers are rapidly reduced under sustained 
ectine, and considerable time is required for a population to recover. 



The black abalone (Haliotis 



form. These organisms frequently are exposed 



where 



may be particularly at risk from human predators because they are easily harvested 
without diving. Individuals of this species reach a maximum size of about 200 mm, 
but few individuals exceed 150 mm. Black abalone have temperature requirements 
similar to red abalone (Ault 1985:15). Black abalone, as in the case of other 



maximum 












55 mm, then growing about 4 mm a year after reaching 90 mm in length (Rickerts 
et al. 1985:245). During early phases of growth, abalones seek the protection of 
rock crevices to avoid predators. Individuals become emergent upon reaching a 
length of 75-100 mm; i.e., they venture onto exposed rock surfaces in search ot 

food (Ault 1985:6). , . . 

The black turban snail (Tegula funebralis) is common to abundant on open rocky 
surfaces of the intertidal habitats that support the black abalone (Morns et al. 
1980:253). The smooth, rounded to conical shell may reach a diameter of 30 mm, 
though many individuals are much smaller. At low tide T. funebrahs is seden- 
tary, and hundreds of individuals often aggregate in rock crevices. Locomotion 
is achieved with a muscular foot in the same fashion as other manne gastropods^ 
Tegula are herbivores that eat many species of algae, including microscopic films 
and kelp fragments (Morris et al. 1980:253). They may reach 20-30 years . o lagc, 
and thus may have a lifespan longer than any other gastropod (Morns. et _al. 
1980:254). Like black abalone, black turban snails are readily exploited by terrestnal 
predators during low tide. 



66 



RAAB Vol. 12, No. 1 



RETHINKING THE LOWLY SHELLFISH 



compared 



emergency or low-yield foods of comparatively minor importance. 
7) argues, for example, that marine environments are generally less 
per unit area than terrestrial environments. He suggests that human 
s tend to relegate marine resources to a lower order of importance if it 



many 



ms 



certain 



misleading in view of the compara 



Shellfish collecting has long been regarded as unimpressive as a gaug 
cultural ' 'advancement." Uhle (1907:31), for example, dismissed the prehisi 
inhabitants of the Emoryville shell mounds in California as representative of 
"lower classes of society" owing to their dependence on shellfish: 

The manner of procuring the essentials of life by collecting shells in itself 
indicates a low form of human existence. In all parts of the world, even 
today, people may be seen by the shore at low water collecting for food 
the shells uncovered by the retreating tide; and although under changed 
conditions of life they raise shellmounds, these people always belong to 



manner a primitive 



sim 



Meehan (1982), in her excellent ethnograph 



points out the many misconceptions and biases that have worked against an 
informed understanding of shellfish economies. Fortunately, the significance of 
shellfish in maritime adaptations recently has received more objective attention. 
Sanger (1988:91), for instance, notes the important role of shellfish in settlement 
patterning and subsistence in the northeastern United States: 

Shellfishing, obviously an important aspect of the Gulf of Maine mari- 
time adaptation, required only the use of hands to pluck mussels from 
rocks, or "pick" clams through their siphon holes... Briefly, shellfish, 
although frequently belittled in the literature as gross contributors to the 
overall caloric component of shell midden sites, may have constituted 
the primary motivation for site selection and abandonment. They con- 
stituted a reliable, if not spectacular, food source, and it would have been 
possible to over-exploit resources in the kinds of intertidal flats commonly 
found associated with habitation sites. A pattern of brief occupation, 
followed by movement to another site, may help to explain the presence 
of nearly 2,000 known shell midden sites an the coast of Maine that have 
survived erosion due to a submerging shoreline. 

Erlandson (1988:107), in evaluating the role of shellfish in the prehistoric 
maritime economy of southern California, makes a similar point: 

... I have tried to counter previous assertions that shellfish exploitation 
is universally inefficient subsistence strategy by demonstrating that, under 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 67 



serve 



yield. This hypothesis has been supported with experimental 



California, and a reexamination 



example from 



exploitation. 



Recent evidence suggests that coastal populations of early Holocene (7000- 
10,000 B.P.) southern California subsisted on a diet in which calories were 
supplied to a large extent by hard seeds, and substantial protein was provided 
by shellfish (Erlandson 1991). 

Relying on both ethnographic and archaeological data, Davidson (1984) 
demonstrates the important role played by shellfishing in the aboriginal 
settlement-subsistence systems around Palliser Bay, New Zealand. Virtually in 
a class by itself, the study by Meehan (1982) of aboriginal shellfishing in Australia 
provides detailed data on the ecology, social organization, nutritional yield, and 
integration with larger economic systems of shellfish economics. 

The result of these investigations is increasing recognition of the distinctive 
and important roles of shellfishing. This recognition should prompt archaeologists 
to go beyond the cursory treatment of shellfish remains that have been 
characteristic of many research efforts of the past: 

... the economics of shellfish exploitation should be evaluated in relation 
to the productivity and reliability of other resource alternatives, including 
factors such as the nutritional role played by various resources, the 
technological investment in resource acquisition, non-food payoffs, the 
availability of various foods to different age and gender groups within 
a society, seasonal limitations on the availability of different resources, 
and differential storability (Erlandson 1989:15). 



SAN CLEMENTE ISLAND 



San Clemente Island, 58 sq mi (148 sq km) in extent, is the southern-most 
of the eight Channel Islands found off the southern California coast (Fig. 1), and 
lies 48 mi from the nearest landfall on the mainland. In historical times the island 
was occupied by the island Gabrielino Indians (Johnson 1988; Bean and Smith 
1978). Although the island's culture history is poorly understood at present, 
recent archaeological studies indicate a human occupation of nearly 10,000 years 



some 



1990b). 



Pacific coast of North America (Raab and Yatsko 1990a 



Miocene 



This process, combined with rising and lowering ocean levels during the 



formed 



(Olmsted 1958) 

that characterizes most 



pport 



st snore iuuajr . nw« ; . 

marine ecosystem characteristic of rocky 



number of gastropod species, including abalones, inhabit 



68 



RAAB 



Vol. 12, No. 1 




San Clemente % ' 

island 



CLEMENTE 
ISL A ND 





5 miles 



5 kilometers 



Contour Inttrval 200 Feet 
Depth Curve in Fathoms 



FIG 



Clemente 



1319, SCII-1325, and the Eel Point site (SCII-43) are indicated. 

Significantly, marine terraces have the greatest number of archaeological sites 
per sq km of any physiographic zone on the island; frequently 200-300 per sq 
km (Yatsko 1987a, 1987b). Most sites are relatively small, discrete shell midden 
sites, some of which have apparent house floor depressions. The site in Fig. ^ 
is characteristic of the small, shallow shell middens found on the coastal terraces 
of San Clemente Island. The Eel Point site complex (SCII-43), covering many acres 
and producing the oldest dated cultural component on the island (9700 B.P.; Sal s 

1988:353-369),is found within this zone. 

Three coastal midden sites, SCII-1318, SCII-1319, and SCII-1325, were 
partially excavated. These sites are located on a high marine terrace, approximate y 
300 m above sea level, on the central west side of the island (Fig. 1). Sites SCII-13 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



69 

















t« 










FIG 



Midden sites on the west shore of San Clemente 
small, soatiallv discrete shell middens. The broken 1 



midde 



typical 
K from 



•"uiuiLUO VJL VJUIC13 111 111C Cll^tl, Ui>w »»».«■— j j * 

existing road. Both sites are black, ashy midden deposits containing abundant 



marine 



with a soil auger, 10 cm in diameter, in order to estimate their subsurface character 
and extent. Based on the auger data, excavation units were placed to obtain 
samples of midden from the deepest to the shallowest portions of the deposits 



at a site. 



pproximately 24 m (north-south) by 16 m 



volume of approximately 41 m 



periphery, two in the deepest midd 



periphery and the two deep pits 



cm 



midden matrix 



70 



RAAB Vol. 12, No. 1 



m in diameter 



volume of approximately 76 m 3 . Four 1 m by 1 m 



within the site; two near the center and deepest portion of the deposits 

the periphery, and one between the center and periphery. Three excava 

nits were due to a deoth of 40 cm. and one unit reached a denth of 30 err 



sterile sediments, for a total of 1.5 m 3 



midden matrix 



Site SCII-1325 was selected to salvage information from a midden damaged 
by past military activities. A military '"foxhole" had been dug into the center of 
the midden, exposing a 60-cm-deep midden stratum. The depth of this site is 
unusual for coastal midden sites in this area. We examined this stratum in the 
area of the site damaged by the foxhole. A portion of the wall of the foxhole was 
cleaned and a vertical profile established. A 1 m by 1 m unit was excavated into 
the midden deposit from the cleaned profile. This site was approximately 30 m 
in diameter, with the excavation unit located near the center of the site. The site 
contains about 212 m 3 of midden. The excavation unit was dug to a depth of 
60 cm, yielding a total volume of 0.6 m 3 of excavated midden matrix. Despite 
efforts to clear away midden deposits disturbed by the excavation of the foxhole, 
variability in compaction of the midden matrix and other factors suggest a reversed 
stratigraphic sequence. 

Two of these sites have been radiocarbon dated. However, these dates pre- 
sent difficulties that cannot be resolved within the scope of this paper. Paired 
samples of charcoal and shell were taken from each excavation level of the two 
sites ("arbitrary" 10-cm levels). Unfortunately, the dates of charcoal and shell 
samples from the same midden proveniences vary greatly; far beyond differences 
one might expect to result from the 14 C "reservoir effect" (Stuiver et al. 1986). 
In radiocarbon years before present these dates are: 

Site SCII-1318: 

Charcoal Dates Shell Dates 

0-10 cm 250 + /-50 (Beta-39148) 1560 + /-60 (Beta-39143) 

10-20 cm 350 + /-50 (Beta-39146) 1777 + /-50 (Beta-39145) 

Site SCII-1319: 

Charcoal Dates Shell Dates 

10-20 cm 240 + /-50 (Beta-39148) 2160 + /-50 (Beta-39147) 

20-30 cm 300 + /-70 (Beta-39150) 2830 + /-50 (Beta-34149) 



No time-diagnostic artifacts were found in the middens. Studies are in progress 
by the author to identify possible sources of errors in radiocarbon dating on 
San Clemente Island, but at present one can only safely conclude that the sites 
are of late Holocene age. 

As noted above, excavation units were dug in 10-cm levels to culturally sterile 
sediments. All excavated midden was passed through 1/8" (5 mm) mesh screen. 
Material collected in the screen was sent to the lab where all recovered midden 
constituents were identified and weighed. The resulting data were entered into 
a computer data base. The sheer volume of shell and other materials to be 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



71 



i recorded imposed a substantial analytical burden. Nearly 480 
were required to analyze and record the 3.3 m3 of midden matrix 



recovered by our excavations. 



spectrum of subsistence remains 



em 



number 



(Haliotis) 



sea mammal bone, and fish bone (Table 1). Most of the abalone shell from the 
middens was fragmentary. All specimens are black abalone (H. cracherodii) , com- 
mon to the island's rocky west shore. Tegula funebralis, the small black turban 
snail inhabiting the intertidal zone of the west shore (Hedgpeth and Hinton 



TABLE 1.— Tabulation by weight (gm) of midden constituents, by stratigraphic 
levels, for sites SCII-1318, SCII-1319, and SCII-1325. 



SCII-1318: 
0-10 cm 
10-20 cm 
20-30 cm 
Totals : 

Site Percentage: 1 

SCII-1319: 
0-10 cm 
10-20 cm 
20-30 cm 



30-40 



cm 



Totals: 

Site Percentage: 1 

SCII-1325: 
0-10 cm 



cm 



10-20 



20-30 cm 
30-40 cm 

cm 



cm 



40-50 
50-60 
Totals: 

Site Percentage: 1 



Haliotis 



151.29 
415.67 
411.83 
978.89 
8.10 



136.00 

1,457.92 

5,051.38 

7,036.14 

13,682.04 

16.63 



157.55 
689 . 65 
965.94 

590.79 

572.64 

68.04 

3, 044 . 61 

8.78 



Tegula 



1,409.70 
2,656.8 
1,579.04 
5,654.61 
65.83 



2,859.44 
17,333.07 
26,837.13 
12,882.62 
59,912.26 
72.82 



1,864.11 
6,533.00 
6,675.00 
5,085.00 
2,885.00 
289.58 

23, 331 . 69 
67.31 



Sea Mammal 



Urchin 



0.07 
0.05 
0.13 
0.25 
0.0029 



6.55 

17.12 

83.42 

79.25 

186.35 

0.23 



1.92 

6.09 

3.10 

11.35 

10.87 

0.83 

34.16 

0.10 



Bone 



1.04 

5.53 

12.77 

19.34 

0.23 



6.31 

15.87 

90.49 

207.44 



0.39 



6.65 
10.28 
75.10 
17.87 

34.35 
00.00 



0.42 



Tegula/ 
Fish Haliotis 



Bone 



4.0 

5.86 

3.13 

12.99 

0.15 



95.84 
69.35 



320.11 190/74 



0.23 



62.01 

37.96 

17.07 

7.83 

0.85 



144.25 148.65 



0.43 



Ratio 



9.3:1 
6.4:1 
3.8:1 



6.27 20.9:1 
19.28 11.9:1 



5.3:1 
1.8:1 
1.8:1 



22.93 11.8:1 



9.5:1 
6.9:1 
8.6:1 
5.0:1 
4.3:1 



lp 



ercentages do not sum to 100% because these figures are not based on the total weight of all 
dietary elements and other midden constituents. 



72 



RAAB Vol. 12, No. 1 



1961:24-25), contributed the largest proportion of shell to the middens. Sea 
urchin remains are of the purple sea urchin (Strongylocentrotus purpuratus) . It may 
be assumed that essentially all mammal bone was derived from sea mammals; 
the island has no indigenous mammals larger than the Channel Islands Fox 
(Urocyon littoralis) . There is no archaeological evidence on San Clemente Island 
indicating that this species was used for food. Fish bones frequently can be 
identified from the spines, vertebrae, syncranial bones, and other elements found 
in the middens. Table 1 presents the weight and percentage of the five classes 
of faunal remains by site and stratigraphic level. 

The high proportion of tegula shell is striking. Table 1 shows that tegula shell 
ranges from 67.31% to 72.82% of all dietary elements by weight. Uncritical assess- 
ment of these figures may lead to unwarranted conclusions. One must approach 
reconstruction of prehistoric diet from midden remains with a number of cau- 
tions in mind. Among these is recognition that the sheer frequency or propor- 
tional representation of a species is not necessarily an accurate index of economic 
importance. Although a midden may contain only a few sea mammal bones, for 
example, the subsistence yield of this resource might have been greater than the 
combined yield of a more abundant faunal class such as shell. 

Reconstruction of dietary yield from archaeological shell samples requires 
attention to several problems. The fragmentary nature of midden shells frequently 
frustrates the task of determining the MNI (minimum number of individuals) 
represented by a shell sample. Frequently, MNI must be estimated on the basis 
of shell weight. Estimates of this type must recognize that after deposition in a 
midden, shell may be transformed by physio-chemical processes. Leaching of 
carbonates may reduce the weight of shells (Kent 1988:9-16), with obvious 
implications for MNI figures based upon weight. 

Tegula and black abalone shells, both fresh and midden specimens, were 
examined to estimate possible leaching effects, and to derive MNI estimates 
based upon shell weight. Comparison of the weight of fresh and midden shells 
of equal size shows that midden shells have lost approximately 3-4% of their 
weight, presumably as a result of the dissolution of carbonates from the latter. 
This estimate should be applied with caution, however, since leaching effects may 
vary with the age of sites. 

A sample of 65 living tegula of various sizes was collected from a tide pool 
on San Clemente Island. These were boiled in water for 10 min to remove these 
small organisms from their shells. Although cooking may result in a slight reduc- 
tion of flesh weight, it was assumed that this loss would be no greater than the 
loss caused by prizing the organisms out of broken shells. Weight of cooked flesh 
averaged 0.47 gm. Average shell weight was 2.59 gm. Applying a 4% reduction 
for shell leaching gives an estimated average weight for archaeological shell 
specimens of 2.49 gm. Using this average, 2,267 individual tegula are represented 
by the total weight of tegula shell in site SCII-1318 (Table 1). At an average flesh 
weight of 0.47 gm per organism, 2,267 tegula would yield 1,065.49 gm (2.34 lbs) 
of meat. Applying the same calculations, site SCIM319 yields 24,061.15 organisms, 
producing 11,308.74 gm (24.92 lbs) of meat; site SCII-1325 yields 9,370.16 
organisms, producing 4,403.97 gm (9.71 lbs) of meat. 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 73 



As in the case of tegula shell, it is possible to estimate the meat yield of a 
sample of abalone shell. This estimate can be made by determining the MN1 
represented by the sample of midden shell, and the meat yield of this number 
of individuals. Once again, living populations afford a partial basis for yield 
estimates. Leighton and Boolootian (1963:237) provide two slightly different 
formulas for estimating the soft-tissue weight of the black abalone from shell length 
measurements. These data derive from studies at Point Dume and the Palos 
Verdes Peninsula, southern California. Both formulas were obtained from regres- 
sion analyses of body weight on shell length. Reflecting slight differences in the 
size range of the two abalone populations, the Point Dume formula yields a frac- 
tionally more conservative estimate; i.e., body weight tends to be slightly less 
in relation to shell size when compared to the Palos Verdes population. Selec- 
ting the more conservative measure, the Point Dume formula is: 

log W = 3.465 log L - 4.668 



(W = soft-tissue weight [gm]; L = shell length [mm]) 



The formula 



lengths from 50-120 mm. It is not clear why this formula was employed instead 
of standard allometric formulas (e.g., Y = aX*>, log Y = b(log X) + log a), but 
Leighton and Boolootian's data offer one method for estimating meat yield from 

shell length. 

Recent studies by the author and others of midden shells and living abalones 
on San Clemente Island are also useful for estimating meat yields. During 1990, 
information was collected on shell length and soft tissue weight of 111 living black 
abalones from three contiguous areas on the island's west shore.2 Abalones of 
all sizes were taken from the intertidal zone in the order in which they were 
encountered. The shell length of each specimen was recorded, along with the 
weight of the foot and viscera (foot weight + viscera weight - total soft tissue 
weight). Information was also collected on abalone shell length and weight from 

middens. 

Data from living specimens offer only a partial solution to the problem of 
fragmentary shells in middens. Estimates of abalone meat yield from midden 
remains involve several factors. Shell length data offer an approach to establishing 
MNI counts. Black abalone shells excavated from the sites discussed here ranged 
from about 50-130 mm, although few shells exceeded 80 mm. Unfortunately, no 
systematic measurements of shell lengths were made during excavation, gently 
collprtPH *hM Hafa hnwpvpr afford a means of characterizing abalone shell length 



midden 



Measureme 



its or wnoie aDaiontr hru <«c «»«*• - - 

™„ «. „« ™»td. Abalone measurements were taken recently at J™ midden 
sites, SCII-315 and SCII-310, on the west shore of San Clemente Island . At site 
SA61 (SCII-315), 51 whole specimens were recovered and measwed, with a mean 
length of 59.8 mm and a standard deviation of 18.7 mm. At site SCII-310, 86 whole 
specimens were recovered with a mean length of 69.2 mm and a standard devia- 
tion of 14 mm. A third sample of shells was obtained from a refuse deposit within 



74 



RAAB Vol. 12, No. 1 



a prehistoric pithouse at the Nursery site (SCII-1215; Sails and Raab 1991). 
This deposit yielded 111 shells with a mean length of 75.2 mm and a standard 
deviation of 18.3 mm. These three archaeological shell samples have a modal 
length of 70-80 mm. 

For purposes of estimating meat yield of prehistoric specimens, a modal shell 
length of 75 mm seems reasonable. Using the Leighton and Boolootian (1963:237) 
formula presented earlier, a meat weight of 67.47 gm is obtained for a shell length 
of 75 mm. The author's study of living black abalone, however, produced a 
somewhat different result: average soft tissue weight for specimens ranging 
70-79 mm (n = 6) was 40.92 gm. It may be appropriate, in this case, to use 40.92 
gm as the average soft tissue weight of an abalone with a shell 75 mm long. 
This figure is derived empirically from an extant local population, 
formula presented by Leighton and Boolootian may tend to overestimate the 
tissue weight of specimens below 100-120 mm. An estimate of total meat 
may be obtained by multiplying the 40.92 gm average body weight bj 
estimated MM represented by a shell sample. 

estimate MNI, it is necessary to know the weight of a 75 mm 



and the 



shell. Fifty-five abalone shells were 



from deflating 



midden deposits at the Eel Point site (SCII-43) in order to obtain information 



weight characteristics. Shells 70-80 mm 



gm 



A weight of 25 gm was therefore used to estimate the MNI represented in 
Table 1, and the number of specimens thus derived were multiplied by 40.92 gm 
(soft-tissue weight). Estimates of total meat yield for turban snails and abalone 
for each of the levels of the three midden sites are presented in Table 2. 

Another factor to be considered here is whether the whole organism was eaten 
or, like modern consumers of abalone, prehistoric peoples ate only the muscular 
foot. Clearly, total dietary yield would vary, depending on the portion eaten. The 



present estimates 



consumptio 



tissue. I am aware of at least one ethnographic instance in which all soft tissue 
was apparently consumed (Meehan 1982:5-6). There appears to be no nutritional 
basis for ruling out such a practice. 

On the basis of meat yield, abalones appear to be the key shellfish compo- 
nent of the middens, although this conclusion might not be apparent from either 
casual observation of the middens themselves or examining the high percentages 
of tegula shell in Table 1 . Of resources gathered in the intertidal zone, abalones 
have the greatest subsistence yield. Tegula were being intensively collected as 
well, making a secondary subsistence contribution. The importance of abalone 
procurement within the subsistence base may be examined in another fashion. 
A factor analysis (Systat statistical program, V. 3.0) of data presented in Table 1 
creates a statistical model in which it is possible to estimate the influence of each 
faunal constituent on each of the other constituents. This procedure is a princi- 
pal components analysis, which determines the total amount of variance explained 
by each of the faunal constituents (components) within Table 1. Variance estimates 
are derived from regression coefficients for all variables in Table 1 . This analysis 
is helpful in determining which dietary components influence the values of others 
within the matrix. The results of this analysis are that the abalone component 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 75 





TABLE 2.- 


-Estimated total meat 


yield (gm) for Tegula and Haliotis. 






Site SCII-1318: 


Haliotis 


Tegula 




0-10 cm 




247.63 


266.00 




10-20 cm 




680 . 37 


501.49 




20-30 cm 




672 . 72 


298 . 05 




Totals: 




1,600.72 


1,065.54 




Site SCII-1319: 








0-10 cm 




222 . 60 


539.73 




10-20 cm 




2,386.32 


3,271.70 




20-30 cm 




8,268.10 


5,065.64 




30-40 cm 




11,516.75 


2,431.66 




Totals: 




22, 393 . 77 


11,308.73 




Site SCII-1325: 








0-10 cm 




257.88 


351.86 




10-20 cm 




1,128.82 


1,233.14 




20-30 cm 




1,581.05 


1,259.94 




30-40 cm 




967.00 


959.82 




40-50 cm 




937.30 


544.55 




50-60 cm 




111.37 


54.55 




Totals: 




4,983.42 


4,403.86 



tegul 



(followed 



greatest 



r words, variation in the contents of the middens 
with the amount of abalone shell present. These dat 
ng was particularly important in structuring the 



lecting pattern. 

It is hardly surprising that abalones and tegula account for mo« 
in Table 1, given the fractional representation of the remaining 
tuents. It is interesting however, that abalones account for tl 



for 



These data suggest that dietary remains from 



samples examii 
in the middens 



seem to play a pivotal role in structuring the subsistence pattern 



estimated meat 



OPTIMAL FORAGING 
The observed data conform well to an optimal foraging model of subsistence 



76 



RAAB Vol. 12, No. 1 



The range of foods eaten by different populations of animals or human 
hunter-gatherers depends both upon the "value" of available, edible 
resources . . ."Value" may be defined as the net energy yield of a resource 
per unit of "handling time;" for human hunter-gatherers, the latter 
includes time devoted to the capture, retrieval and processing of different 
foods. The result is that most foragers take an optimal range of resources, 
or optimal dietary breadth . . . When is a particular resource included in 
(added to) the optimal dietary breadth? This appears to depend pri- 
marily on the "value" of that item, and both the abundance and the value 
of more valuable resources . . . Thus, the value (handling time) of an item 
is the major factor, and abundance a secondary factor, in determining 
the resources "selected" to constitute the optimal diet (Yesner 1981:150). 



Moreover: 

Large, carnivorous foragers, such as humans, tend to exploit their environ- 
ments in a "fine-grained" fashion, encountering and exploiting resources 
in the proportions in which they actually occur . . . Thus, for those species 
that are harvested, an optimal forager will take them in amounts represen- 
tative of the biomass of that species in the (local) environment. In other 
words, while the spectrum of species eaten may not be a representative 
sample of all species available, among those that are eaten, harvested 
biomass should be proportional to that in the natural environment (Yesner 
1981:150). 

Although the present discussion focuses on the five midden constituents 
presented in Table 1, the middens actually contain a broad spectrum of food 
remains from the intertidal zone, albeit at quite low frequencies. We find remains 
of limpets, chitons, crabs, and birds, for example. The middens appear to con- 
tain essentially every edible tide pool species likely to have been taken by a "fine- 
grained" mode of foraging. For this reason it seems likely that the relative 
proportions of abalones and tegula in the middens reflect the availability of these 
species in the tide pool environment, an observation reinforced by the present 
intertidal ecology. The foraging strategy employed by the prehistoric tidepool 
gleaners appears to reflect the fact that, of all species available, only abalones 
and tegula could be collected in sufficient quantities to meet dietary needs. 

The abalone is clearly the single most "valuable" species represented in the 
middens; i.e., this species would yield the greatest subsistence return in relation 
to the "handling time" involved in procurement. The black abalone, as discussed 
earlier, is a "meat package" of substantial size. This species, being an intertidal 
form, is also relatively easy to exploit; they are readily collected on rocks expose 
during low tides. This food resource is also easily exhausted. Even a small group 
of foragers, given daily collecting trips, could quickly strip a sizable section o 
the shore of abalone. This suggests that the optimal dietary breadth was dynamic; 
i.e., was determined not only in relation to the types of food resources encountere 
in the tide pools, but also by the impact of the foragers themselves. 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 77 



Any decrease in the availability of abalone would likely change the value of 



m 



food items; on this basis, their dietary value may appear to be minimal at best. 
On the basis of the optimal foraging model, however, their value as a food item 
would be determined in relation to their abundance, the availability of food items 
of higher value, and their handling cost. The imposing presence of tegula in the 
middens clearly indicates that they were abundant. In the event of an insuffi- 
cient supply of higher value shellfish (abalone), their abundance alone may recom- 
mend tegula as a food item. The ease with which tegula may be obtained reduces 
their handling time, thus increasing their relative value. These snails are easily 
collected by hand in tide pools during low tide. Ease of procurement also means 
that most members of the population could participate in economic production 
without special skills or equipment. Significantly, children and the aged, persons 
otherwise marginal to the economy, can make significant contributions to the diet 
by collecting shellfish. Although nutritional data on tegula are lacking, as algae- 
consuming gastropods, it seems reasonable to assume that they would yield nutri- 
tional benefits on par with abalone (c.f. Erlandson 1988:104). 

But what about the handling time involved in extracting food from such a 
small organism? At first this factor might seem to weigh against the dietary value 
of tegula. An interesting comparison can be made here to research by Yesner 



middens 



amon 



exam 



x ^x V.AUIHJ/1V., """V. ,_vxiwi »; r » r - - *I 

relatively abundant on the strand flat, comprising about twenty per cent 
of the shellfish biomass, they are practically non-existent in the middens, 
and there is no ethnographic evidence that they were ever significant 



items 



These species probably fall outside the optimal 



time 



smaller 



picks would have been required to retrieve the individually small 



amount of (low calorie) meat from 



From this perspective, it may be surpris] 
The midden evidence from San Clemente 



small 



from the middens has been crushed 



tegul 



The 



organisms are freed from 



handling costs. 



The intertidal ecosvstem 



ine intertiaal ecosystem or ^>dii ucinwuv ~~ 

certain resource potentials for shellfish exploitation. The fashion in wlnchtha 

ecosystem was utilized conforms well to an optimal foraging mode of '«***"«; 
Black abalone exploitation was the single most valuable component of the shel foh 
economy. Since this resource was easily depleted, however, the dietary breadth 



78 



RAAB Vol. 12, No. 1 



was necessarily dynamic, with black turban snails collected in large quantities 
to supplement dwindling supplies of abalones. This pattern is indicated in the 
shifting ratios of tegula to abalone meat yield from the lower to the upper levels 
of the middens (Table 1). 

If the conclusions presented here are correct, other aspects of the prehistoric 
ecosystem come into focus. Intensive collecting for food items within the inter- 
tidal zone, even by small human populations, easily has the capacity to exhaust 
the supply of species such as abalone and tegula. As noted earlier, Yatsko (1987a) 
has documented hundreds of shell midden sites of the kind described here on 
the coastal terraces of San Clemente Island. Frequent residential moves are 
perhaps the only means by which a hunting-and-gathering society can cope with 
the easily depleted resources of the intertidal zone on San Clemente Island. High 
mobility of this type could account for the thousands of small midden sites on 
the island's west shore. This is a settlement dynamic similar to that described 
by Sanger (1988:91). 

The intensity of prehistoric abalone collecting on San Clemente Island is 
suggested by the modal size of midden specimens, which ranged between 70 and 
80 mm. As noted earlier, abalones are emergent when they reach a length of 
75-100 mm (Ault 1985:6); i.e., they leave the protection of rock crevices to forage 
for food on exposed surfaces. The midden specimens suggest that few 
postemergent abalones escaped human predators. 



CONCLUSIONS 



Detailed analysis of aboriginal shell middens on San Clemente Island indicates 
that exploitation of shellfish species conforms well to predictions derived from 
an optimal foraging model of subsistence. Although these middens are composed 
largely of tegula shells, estimates of meat yield indicate that the black abalone 
provided the greatest proportion of food. A factor analysis shows that abalone 
shell weights best explain variance in the composition of the total shell 
assemblages. An aboriginal strategy of shellfish collecting is proposed in which 
an optimal dietary breadth responds dynamically to rapid depletion of abalones 
(highest value food source) through exploitation of tegula, a second rank but 
relatively low-cost food source. 

NOTES 

iThe research described here is part of a five-year cooperative research agreement between 
the Center for Public Archaeology, California State University, Northridge and the 
Natural Resources Office, North Island Naval Air Station, San Diego, California. San 
Clemente Island has been a Naval reservation since 1934. Prehistoric settlement pat- 
terning and human ecology are major topics addressed within a research design (Raab 
and Yatsko 1990a) being implemented through site surveys, specialized site testing 
programs, and archaeological field schools. 

2The abalone studies were conducted under terms of a scientific collecting permit issued 
by the California State Department of Fish and Game. Field reports on several hundred 
living abalones on San Clemente Island were provided by Mr. Peter Haaker of the 
California State Department of Fish and Game. 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



79 



ACKNOWLEDGEMENTS 



The research reported here would not have been possible without the assistance of many 
people. The generous support of the United States Navy is much appreciated, particu- 
larly the efforts of Commander John Gana, Officer in Charge, San Clemente Island, and 
Commander John B. Duke, Staff Civil Engineer, NAS North Island. The encouragement 
and active assistance of Jan Larson and Andy Yatsko, Natural Resources Office, Naval 
Air Station, North Island, were of tremendous benefit. Thanks are also extended to the 
students and staff of the San Clemente Island archaeological field schools who helped 
gather the data. Dr. Roy Sails and Dana Bleitz deserve thanks for sharing their knowledge 
of prehistoric exploitation of faunal resources in the Channel Islands. Peter Haaker of the 
California Department of Fish and Game has my gratitude for sharing information on 
abalone populations. Dr. Jon Erlandson and anonymous reviewers offered valuable 
assistance in shaping this paper. I thank Dr. Antonio Gilman for his Spanish language 
rendition of the paper's abstract. Any errors of fact or interpretation are my own. 



LITERATURE CITED 



AULT, JERALD S. 1985. Black, green and 
red abalones. U.S. Department of the 
Interior, U.S. Army Corps of Engineers, 
Biological Report 82(11.32). 

BEAN, LOWELL W. r and CHARLES R. 
SMITH. 1978. Gabrielino. Pp. 538-549 
in Handbook of North American Indians, 
Volume 8, California. Robert F. Heizer 
(editor). Smithsonian Institution, Wash- 
ington, D.C. 

COX, KEITH W. 1960. Review of the abalone 
in California. California Fish and Game 

46(4):381-406. 

DAVIDSON, JANET. 1984. The Prehistory 
of New Zealand. Longman-Paul, Auck- 
land. 

ERLANDSON, JON M. 1988. The role of 
shellfish in prehistoric economies: A pro- 
tein perspective. American Antiquity 
53:102-109. 

1991. Shellfish and seeds as 

optimal resources: Early Holocene sub- 
sistence on the Santa Barbara coast. Pp. 
89-100 in Hunter-gatherers of early Holo- 
cene coastal California. Jon M. Erlandson 
and Roger H. Colten (editors). Perspec- 
tives in California Archaeology Vol. 1. 
Institute of Archaeology, University of 

California-Los Angeles. 
GLASSOW, MICHAEL A. 1977. An archa- 
eological overview of the northern Chan- 
nel Islands, California. Report prepared 
for the Western Archaeological Center, 
United States Park Service, Tucson. 

LARRY R. WILCOXON, and 



JON M. ERLANDSON. 1988. Cultural 
and environmental change during the 



Early Period of Santa Barbara Channel 
Prehistory. Pp. 64-77 in The Archaeology 
of Prehistoric Coastlines. Geoff N. Bailey 
and Jon E. Parkington (editors). Cam- 
bridge University Press, London. 
HEDGPETH, JOEL and SAM HINTON. 
1961. Common Seashore Life of South- 
ern California. Naturegraph Publishers. 



Happy Camp, California. 



The 



— — ^_- w » — - — — — j 

Book. Naturegraph Publishers, Happy 
Camp, California. 

JOHNSON, JOHN R. 1988. The people of 
Quinquina: San Clemente Island's origi- 
nal inhabitants as described in ethno- 
historic documents. Report prepared for 
the Natural Resources Office, Naval Air 
Station, North Island, San Diego. 

KENT, BRETTON W. 1988. Making Dead 
Oysters Talk, Techniques for Analyzing 
Oysters from Archaeological Sites. Mary- 
land Historical Trust, Historic Saint 
Mary's City, Jefferson Patterson Park and 
Museum, Saint Mary's City. 

LEIGHTON, DAVID and RICHARD A. 
BOOLOOTIAN. 1963. Diet and growth 
in the black abalone, Haliotis cracherodii. 
Ecology 44:227-238. 

MEEHAN, BETTY. 1982. Shellbed to Shell 
Midden. Australian Institute of Aborigi- 
nal Studies, Canberra. 

MORRIS 



•.• 



Pacific Coast Shells. Houghton Mifflin, 



Boston. 



MORRIS, ROBERT H., DONALD P. AB- 



BOTT 



HADERLIE. 



1980. Intertidal Invertebrates of Califor- 



80 



RAAB 



Vol. 12, No. 1 



LITERATURE CITED (continued) 



nia. Stanford University Press, Palo Alto. 

OLMSTEAD, F.H. 1958. Geological recon- 
naissance of San Gemente Island, Cali- 
fornia. Geological Survey Bulletin 1071- 
B. United States Printing Office, Wash- 
ington, D.C. 

OSBORN, ALAN J. 1977. Strandloopers 
mermaids, and other fairy tales: Ecologi- 
cal determinants of marine resource 
utilization-The Peruvian case. Pp. 157- 
205 in For Theory Building in Archa- 
eology. Lewis R. Binford (editor). Aca- 
demic Press, New York. 

PIANKA, ERIC R. 1974. Evolutionary Ecol- 
ogy. Harper and Row, New York. 

RAAB, L. MARK and ANDREW YATSKO. 
1990a. Prehistoric human ecology of 
Quinquina: A research design for archa- 
eological studies of San Clemente Island, 
southern California Pacific Coast Archa- 
eological Society Quarterly 26 10-37. 

1990b. Ancient maritime 



Lompoc 



adaptations of the California Bight: A 
perspective from San Clemente Island, 
California, in Essays on the prehistory of 
maritime California. Terry Jones (editor) 
University of California, Davis, Center 
for Archaeological Research Vol. 10. 
(In Press) 

RICKETTS, EDWARD F., JACK CALVIN 
and JOEL W. HEDGPETH. 1985. Bet- 
ween Pacific Tides, 5th ed., revised by 
David W. Philips. Stanford University 
Press, Palo Alto. 

SALLS, ROY A- 1988. Prehistoric fisheries 
of the California Bight. Unpublished 
Ph.D. dissertation, Department of Anth- 
ropology, University of California, Los 



Angeles. 



, and L. MARK RAAB. 1991. 



Prehistoric residential structures of coas- 
tal southern California. Manuscript on 



fornia 



1988 



turns in the Gulf of Maine. Archaeology 
of Eastern North America 16:81-99. 
STUIVER, MINZE, G W. PEARSON, and 
TOM BRAZIUNAS 1986. Radiocarbon 
age calibration of marine samples back 
to 9000 cal yr bp. Radiocarbon 28:980- 

1021. 
UHLE, MAX. 1907. The Emeryville shell 
mound. University of California Publica- 



tions in 



haeology 



>logy 7:1-84 



WALKER 



SNETMKAMP. 1984. Archaeological in- 
vestigations on San Miguel Island-1982: 
Prehistoric adaptation to the marine 
environment ( Final report) Univer ty of 
California-Santa Barbara, Office of Public 
Archaeology, Social Process Research 

Institute 

rSKO, ANDREW. 1987a. Reassessing 

archaeological site density at San Cle- 
mente Island. Paper presented at the 
Third California Islands Symposium, 



Barba 



torv 



1987b. Topographic/geomor- 

phic zones at San Clemente Island for 
anal iis of archaeological site distri- 
butions. Manuscript on file, Natural 
Resources Office, Naval Air Station, 
North Island, San Diego 
YESNER, DAVID R. 1981 Archeological 
applications of optima! foraging theory: 
Harvest strategies of Aleut hunter- 
gatherers. Vp 14H-170 in Hunter-Gather- 
er Foraging Strategies Bruce Winter- 
halder and Eric A. Smith (editors). Uni- 
versity of Chicaeo Press, Chicago. 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 81 



BOOK REVIEW 



Islands, Plants, and Polynesians: An Introduction to Polynesian Ethnobotany. 

Paul Alan Cox and Sandra Anne Banack (editors). Portland, Oregon: Dios- 
corides Press, 1991. Pp. 240, 16 photos. $34.95 (cloth). ISBN 0-931146-18-6. 



Islands, Plants and Polynesians is a collection of papers originally given at a 
symposium entitled "Plants and Man in Polynesia" in Laie, Hawaii, in December 
1988. The result is a beautifully produced book containing a wealth of informa- 
tion about Polynesian ethnobotany. Throughout, one appreciates the ingenuity 
the Polynesians have displayed in adapting to their island environments. Espec- 
ially striking are the influences of their remarkable sea voyages in shaping the 
botanical and cultural landscapes of this area. 



circumscribed 



Emph 



limited to, the tropical island ecosystems of this region. The volume begins 
with Raymond Fosberg's introduction to Polynesian plant environments, in which 
he describes their physical characteristics and reconstructs pre-Polynesian, pre- 
European, and post-European landscapes and biota. Sandra Anne Banack follows 



materials used in Polynesian voyaging, from 



monial 



documents 



European contact are detailed by Arthur Whistler 



and another 12 which were possibly introduced intentionally. Douglas Yen 
addresses the origins of Polynesian cultivars and agricultural systems, discuss- 
ing the influences of Southeast Asia and New Guinea on Polynesian agricultural 



exam 



deducing plant origins and mi 



vernacular names for the sweet potato. Patrick Kirch characterizes aboriginal 



ms 



medicine 



through time. The Polynesian uses of seaweeds are detaiiea Dy isaoe 
Paul Cox provides an insightful discussion of Polynesian herbal 
touching not only on some important plant species utilized, but examining Polyne- 
sian theories of disease causation, the role of healers in Polynesian society and 
the evidence for and against the autochthonous origins of Polynesian herbalism 
The ethnobotany of Kava (Piper methysticum) , an inebrient used throughout 
Polynesia, is discussed in detail by Vincent Lebot. Finally, Diane Ragone gives 
an account of the distribution, uses, and preservation methods for breadfruit 

(Artocarpus altilis) in the south Pacific. 

There is much here to interest both the botanist and anthropologist Scien- 
tific names as well as local names are used accurately throughout the text along 
with the plant family, a welcome practice that facilitates accurate ^tification 
and comparison of botanical entities. An index to scientific names and Polyne- 

words is given. The book is handsomely produced, reasonably pricec and 
easure to read. I found very few errors. Many of the papers emphasize the 




82 



BOOK REVIEW Vol. 12, No. 1 



need to document the ethnobotanical lore of Polynesia before both the plants and 
the knowledge of how to use them are irrevocably lost. This book is an out- 
standing contribution that adds greatly to our knowledge of Polynesian plants 
and peoples, and points the way toward new directions for study in Polynesian 



ethnobotany. 



Lynn Bohs 

Department of Biology 
University of Utah 
Salt Lake City, Utah 84112 



/. Ethnobiol. 12(1):83-115 



Summer 1992 



TWO PREHISTORIC PUEBLOAN AVIFAUNAS 
FROM THE PECOS VALLEY, SOUTHEASTERN NEW MEXICO 



STEVEN D. EMSLIE 

Environmental Studies Department 

University of California 
Santa Cruz, CA 95064 



JOHN D. SPETH 

Museum of Anthropology 
University of Michigan 
Ann Arbor, MI 48109 



REGGE N. WISEMAN 

Office of Archaeological Studies 

Museum of New Mexico 

Santa Fe, NM 87504 



ABSTRACT. -We 



historic sites in the Pecos Valley, Chaves County, southeastern New Mexico. 



Th 



site (LA-25277), a small pitroom community of only three or four structures, 



with 



taxa 



avifaunas, including the easternmost 



Macaw 



during mi 



in winter, suggesting that exploitation of many aquatic species (e.g., geese and 
ducks) was occurring during the winter months. Comparison of these avifaunas 
indicates site-specific differences in the use of birds, with a greater emphasis on 



The 



prehistoric Pueblo Indians practiced site-specific hunting specializations for 
particular species or groups of birds. Trade (directly or indirectly) with 
Mesoamerica is indicated by the presence of the Scarlet Macaw and possibly also 
by the presence of the cardinal. 

RESUMEN.-Presentamos la identificacion de restos de avifauna de dos sitios 
prehisto'ricos tardios en el Valle de Pecos, Condado de Chaves, al sureste de 
Nuevo Me'xico. El Sitio Henderson (LA-1549), un pueblo con 50 a 70 cuartos, 
y el Sitio Rocky Arroyo (LA-25277), una pequeha comunidad de cuartos ; foso con 



estructuras 



1325 



los menos 48 taxa (glneros y especies) estan representados en estas avitaunas 
incluyendo el registro mas oriental conocido de la guacamaya escar ata (Ara macao) 

. r , ^ ^ t, tt i '~a„ mrohi ci-nrim Todos los demas taxa 



84 



WISEMAN Vol. 12, No. 1 



Mexico 



invierno 



que el aprovechamiento de varias especies acuaticas (gansos y patos) se hacia 
durante los meses invernales. La comparacio'n entre estas avifaunas indica 
diferencias especihcas a cada sitio en el uso de aves, con un mayor enfasis en 
los taxa acuaticos en Rocky Arroyo. Las avifaunas tambie'n apoyan la hipotesis 
de que los indfgenas Pueblo prehispanicos practicaban especializaciones de caza, 
especificas a cada sitio, para especies o grupos de aves particulares. El comercio 
(directo o indirecto) con Mesoamerica es indicado por la presencia de la guacamaya 
escarlata, y posiblemente tambien por la presencia del cardenal. 

RESUME.— Nous pre'sentons 1' identification de restes d'oiseaux venant de deux 
sites prehistoriques de la vallee du Pecos (comte de Chaves, au Sud-Est du 
Nouveau-Mexique) . Le site d'Henderson (LA-1549), un pueblo de 50 a 70 pieces, 
et celui de Rocky Arroyo (LA-25277), une petite communaute' de trois ou quatre 
structures semisouterraines, etaient a peu pres contemporains, leur principale 
periode d' occupation se situant tres probablement entre 1250-75 et 1325 ap 
J.C. Au moins 48 genres et especes sont represente's parmi ces oiseaux, y-compris 
l'occurence la plus orientale connue dans le Sud-Ouest prehistorique pour le 
Ara ecarlate (Ara macao) . Tous les autres types se trouvent a present au sud 
du Nouveau-Mexique, mais quelques-uns seulement pendanHes migrations 
et en hiver, ce qui suggere que 1'exploitation de nombreuses especes aquatiques 
(oies et canards) avait lieu pendant les mois d'hiver. La comparaison de ces 
restes d'oiseaux indique des differences specifiques a chaque site dans l'utili- 
sation de ceux-ci, en particulier un usage plus intensif des types aquatiques a 
Rocky Arroyo, et permet de supporter l'hypothese selon laquelle les Indiens 
Pueblos prehistoriques pratiquaient une chasse specialised de groupes et d'especes 
d'oiseaux specifiques a chaque site. Le commerce (direct ou indirect) avec le 
Mexique est atteste par la presence du Ara ecarlate, et peut-etre par celle du 



cardinal. 



INTRODUCTION 



The Henderson (LA-1549) and Rocky Arroyo (LA-25277) sites are two late 
prehistoric (ca! A.D. 1250-1400) puebloan villages located about 8 km apart in 
the Hondo River drainage, a major western tributary of the Pecos River, Chaves 
County, southeastern New Mexico (Fig. 1). The Henderson site, partly excavated 
by personnel from the Museum of Anthropology of the University of Michigan 
in 1980 and 1981, is situated on a low limestone ridge that flanks the south edge 
of the Hondo where this drainage first enters the alluvial flats of the Pecos Valley. 
The elevation of the Henderson site is about 1,186 m. The Rocky Arroyo site, 
at an elevation of about 1,133 m, is located in the alluvial flats of the Pecos Valley 
almost due east of Henderson, about 100 m east of the Rocky Arroyo drainage 
and approximately 1 km upstream (south) from the confluence of this sma 
intermittent drainage with the old channel of the Hondo.* Rock Arroyo was 
excavated almost in its entirety during the 1970s by amateurs belonging to the 
Chaves County Archaeological Society (CCAS). Small-scale salvage excavations 
in one of the structures at Rocky Arroyo were conducted there in 1980 by one 
of us (RNW). 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



85 




FIG. 1.— Location of the Henderson site (LA-1549) and Rocky Arroyo site 
(LA-25277), Chaves County (shown by dashed line), southeastern New Mexico. 



The 



century by overgrazing 



damming 



commercial 



growing city of Ros well . These 



tion cover, and favored increases in more drought-resistant species. Today, 



limestone 



gramas (black, Bouteloua eriopoda 



while 



(Sporobolus sp.) occur along the margins of the normally dry 
and Rocky Arroyo drainages. Other plants common tod 
broom snakeweed (Gutierrezia sarothrae), prickly pear and 



spp.), grease wood (Sarcobatus sp.), four-wing salt bi 
mesquite (Prosopis spp.). Very few trees grow near 



most 



Wild species are found primarily near the Henderson site, where they grow m 
a narrow band along the channel of the Hondo. Among the more common trees 
are cottonwood (Populus sp.), hackberry (Celtis sp.), wild walnut (Juglans sp.), 
and salt cedar (Tamarix sp.), the last introduced historically to control erosion. 



WISEMAN Vol. 12, No. 1 



• it 



The climate of the Roswell area today is semiarid. Roswell rec« 
mm of rain annually (Houghton 1974:802). Winters are relativel; 
oeriod from November through March receiving about 58 mm 



mm 



during the seven months from April to October. Rains come in two distort 
periods. The first is in May, followed by a slight decline in June. The principal 
rainy season, often characterized by intense thunderstorms, occurs from July 
through September; nearly 45% of the annual precipitation falls during these 

three months. . r 

Summers in Roswell are warm, with average daily maxima exceeding 32 L 

(90°F) from June through August. Winters are mild, with average daily highs, 

even in January, of 12.8°C (55. 1°F). Average minimum daily temperatures drop 

below freezing from mid-November through mid-March, but rarely drop to K 

The average length of the frost-free season, 206 days, extends from 7 April to 

30 October (Von Eschen 1961:51). 

The excavation and salvage operations at the Henderson and Rocky Arroyo 
sites yielded a comparatively large sample of avian remains: 548 specimens 
representing at least 120 individuals of at least 48 taxa. We present identifications 
of these avifaunas, which provide information on the use of birds by the late 
prehistoric inhabitants of this fascinating but poorly known region that borders 
tv,p P*etPrn pHctp of the Greater Southwest and western Great Plains. 



SITE DESCRIPTION AND METHODS 

The Henderson site. -The Henderson site is an E-shaped adobe pueblo of 50-70 
large, rectangular, single-story rooms (Fig. 2). The room blocks are arran J 
with the longest or main bar of the "E" oriented roughly 60 degrees wesr 
true north and the shorter bars (referred to henceforth as the east, center ' g 
west bars) extending to the south (i.e., away from the Hondo). The open sp 



,o t v» between the shorter bars form small plaza areas, designated 
west plazas. No evidence of a kiva has been found in either plaz 



and 



rooms tested to date shows evidence of specialized 



functions 



».Ug 

Most of the excavations at Henderson have been confined to r0 ^ mS ^ Il and 
east and center bars, and to the south end of the east plaza (Fig. A _^g • ' e(J 
Fig. 4). With the notable exception of a series of subfloor burials, wMc y ^ 
a diversity of grave accompaniments (see Rocek and Speth 1986), ^J ures 
artifacts were found in situ on room floors or cached in pits or othe ^ ar _ 
Instead, most archaeological materials (e.g., lithics, sherds, animal tfon ^ ^ 
coal) were found randomly dispersed throughout room fill. Only two ^ ^ 

midden deposits were encountered, one nearly filling room dep0S its 
•, the other toward the south end of the east plaza. These tw v 



produced most of the animal bones at Henderson, including many 



of the bird 



remains reported here. . - t app ears 

The midden in the east plaza deserves further comment. This dep » r 



[he midden in tne east plaza deserves runner euiiunau. ..- - - . ^ cen i 
present the remains of a huge roasting complex, located in or a 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



87 



o 

CD 




CONTOUR INTERVAL 25 CM 
SITE ELEVATION CA 1 185 M (3890 i 



490 



515 



FIG. 2.-Map of Henderson site (LA-1549) showing location of principal exca- 



vation units. 



Pi 



While 



proximity is indicated by the large quantities of burned and fractured limestone 



limestone 



animal 
Much 



in the plaza, particularly the butchering and roasting of bison, but domestic 
trash from adjacent room blocks also appears to have been dumped into the 
gully as evidenced by large quantities of broken pottery and lithics, and debris 
from the manufacture of freshwater mussel shell ornaments. 



77? 



In contrast to the Henderson site, Rocky Arroyo con- 
three large, deep rectangular pitrooms, irregularly spaced adja- 
midden-filled area, with no evidence of above-ground structures 
>\ The trash deoosits and associated pitroom structures at Rocky 



almost com 



The exca- 



88 



EMSLIE, SPETH & WISEMAN 



Vol. 12, No. 1 



vators appear to have been systematic in their search for artifacts, screening 
much of the fill through quarter-inch mesh. Their excavations focused on three 
structures, two of which they emptied completely; they also excavated most 
nf thp associated midden deposits. 



524N 



523N 



522N 



521N 



514N 



513N 



512N 



511N 



510N 




519N 



518N 



517N 



516N 



p 11 d ♦ l ^- y Beta- 14069 

1406J^ X VHearth 
-X»l F.43 ) 



'- - -Y\L-^'Ash Pit) 





14072 



F.45 



HENDERSON SITE 
(LA- 1549) 



- TRENCH F (CENTER BAR) 



KEY 





Post 
Limestone Slab 



F. Feature 



.5 1 




Meters 



530E 



531E 



533E 



534E 



of center bar at Henderson 



rooms and features. 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



89 



HENDERSON SITE (LA- 1549) 
TRENCH A (EAST BAR) 



KEY 




Post 

Limestone Slab 




F. Feature 



529N 




528N 



527N 



526N 



525N 



524N 



523N 



522N 



521N 



561E 



FIG. 4 

and features. 



Henderso 



rooms 



material 
remains/as well as many human bones (mos 



unknown, but faunal remains, as wen as many uuuiai. w «~ y j 

and phalanges), lithics, sherds, and other debris were discarded and left m 
piles on the surface of the site. The excavators placed the bones into lar^m^ 
tins, which they left near the structures where, presumably, the bones had been 
recovered; two of us (RNW and JDS) salvaged these materials and treated I them 
with a polyvinyl-acetate preservative to prevent the bones from disintegrating. 



90 



WISEMAN 



Vol. 12, No. 1 




FIG. 5.— Sketch map of Rocky Arroyo site (LA-25277) showing approximate 
location of pitrooms and general midden deposits. 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 91 



The biases in this faunal sample are difficult to assess. We are unable to 
determine with any certainty which bones came from structures and which 
came from extramural midden deposits, nor can we tell whether a tin contained 
bones from the structure nearest to it or from more distant ones as well. In 
addition, the comparatively small number of bones of small mammals and birds 
(e.g., rabbits, rodents, and passerines) suggests that either much of the deposits 
was not screened or that only larger bones were collected from the screens. 

Fortunately, one of the three pithouses, Structure or Locus 2, was not totally 
emptied. A large wedge of undisturbed deposit was excavated by one of us (RNW) 
and all sediment processed by flotation and fine-screening using a 0.5 mm 
geological sieve. This sample contained thousands of well-preserved fish bones 
(including scales), as well as a wide array of other faunal remains, such as bison 
(Bison sp.), pronghorn (Antilocapra americana), muskrats (Ondatra zibethicus), 
and many rodent, rabbit, tortoise, and bird bones. 



Age and correlation. —Ceramic 



period 



archaeomagnetic dates place the principal occupation at both sites between 
ca. A.D. 1250-1275 and A.D. 1325, with continuing but less intensive occupa- 
tion or intermittent reoccupation of the Henderson site in the late 1300s or 
early 1400s (see Rocek and Speth 1986; Wiseman 1985). 

While the ceramic assemblages from Henderson and Rocky Arroyo are 
virtually indistinguishable and suggest broadly equivalent ages for the two 
sites, the architectural differences between them-pithouses at Rocky Arroyo, 
above-ground adobe structures at Henderson-suggest that Rocky Arroyo may 
be slightly earlier than Henderson. However, in the absence of tree-ring dates 
or other criteria for precisely establishing the dates of the occupations, we 
cannot rule out the possibility that the architectural differences reflect ethnic, 
seasonal, or other factors rather than chronology. 



The faunal remains. —Most faunal remains 



from arbitrary 5 cm 



square grids), with all matrix 



small number of avian specimens were recovered from 



more 




samples were processed from a wide vanety 



information written directly on them 

datum assipnpd an plpvation of 100. ( 



n specimens have complete provenience 
lenoting grid square and depth below a 
m (e £., 513N565E 101.20-101.15). Most 



ones from different units and archaeological 
lot" numbers. These numbers are given for 



Appendix table.2 



from the backdirt piles at Rocky Arroyo were 



mbers based on the 



,uv - u » numDers Dasea on tne arcnutxiiuai w w^.v.. »- 

were found (Fig. 5). Four such loci were identified, three of which (Loc. 1-^ 

__. .. ° ' _ . /-j i--.. ~ t ™^ 1 and ? rould not be 



pitroom structures. Some bone found 



<«e pitroom structures. t>ome Done iuunu ucivv W . 

assigned with confidence to either area and was labeled Loc. 1-2. Salvage exca- 



92 



EMSLIE, SPETH & WISEMAN Vol. 12, No. 1 



vations (by RNW) of the undisturbed deposits in Locus < 
enlarged the total faunal sample, although most material 



and fish bones. These bones were assigned lot numbers to represent their 

stratigraphic provenience. 

All bird bones were identified (by SDE) at the Museum of Vertebrate Zoology, 
University of California, Berkeley. 3 Minimum numbers of individuals (MNI) 
for each taxon were determined by counting the most common element from one 
side and by morphologic comparisons of elements of different sides. The per- 
centage of total number of identifiable specimens (NISP) per taxon was used to 
compare faunal samples by employing a test of equality and arcsine transfor- 
mation to calculate a test statistic (t s ; Sokal and Rohlf 1969). We used this test 
to determine significant differences (p<0.05) in the percentage NISP by Order 
(intersite comparisons) and by skeletal element (intrasite comparisons) from 
Rocky Arroyo and Henderson. Comparisons of the proportions of wing versus 
leg elements by provenience were based on three paired elements for each 
category: radius, ulna, and carpometacarpus as wing elements because these 
bones support the majority of wing feathers, and femur, tibiotarsus, and tarso- 
metatarsus as leg elements. We assumed no differential preservation of wing ver- 



elements 



DISCUSSION 



At least 48 taxa (genera and species) are represented by these avifaunas. 
Except for the macaw (Ara macao), all species are currently found in New Mexico, 
but many are migratory and occur in southern New Mexico primarily in the winter 
or spring and summer seasons (Table 1). The numerous aquatic taxa, especially 
from Rocky Arroyo, indicate the former presence of large rivers, marshes, and/or 
lakes. The Hondo River presently is dry most of the year as a result of damming 
and a lowering of the water table, but was a permanent stream in the late 1800s 
(Rocek and Speth 1986). Several species in the avifauna, including prairie falcon, 
scaled quail, and Chihuahuan raven, occur in open grasslands of the desert 
Southwest. 'Grasses and shrubs are the dominant vegetation of this semiarid 
region of New Mexico today (Rocek and Speth 1986). 

The avifaunas reflect a much richer environment in the past than is present 
in this area of New Mexico today. This environment included a permanent source 
of water, the Hondo River, with associated wetlands and riparian habitats, and 
large, open grasslands. It is possible that some of the birds, particularly the 
aquatic species, were captured some distance away from the villages; if so, 
wetlands along the Pecos River, such as those still seen today in the Bitter Lake 
National Wildlife Refuge and the Bottomless Lakes State Park (both 25-30 km 
east of the villages), would have offered ideal habitats. 

Intersite comparisons. —In addition to architectural differences between the Hender- 
son and Rocky Arroyo sites, the avifaunas also differ (Table 1). Some of these 
differences are due to excavation and recovery biases against smaller bones (and 
hence against smaller species) at Rocky Arroyo. This bias is apparent in com- 
paring the proportion (based on NISP values) of passerines in the Henderson 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 93 



TABLE 1.— Avifauna from the Henderson (LA-1549) and Rocky Arroyo (LA-25277) 
sites, Chaves County, New Mexico, with total number of bones (NISP) and 
minimum number of individuals (MNI, in parentheses) for each taxon. 



Taxon 



Trumpeter Swan (Cygnus buccinator) 
Snow Goose (cf. Chen caerulescens) 
Canada Goose (Branta canadensis) 
Goose (Anserini, indet.) 
Mallard (Anas platyrhynchos) 
Duck (Anas sp.) 

sp 



** 



Hawk (Buteo sp.) 



h rysaetos) 



Merlin 



Peregrine Falcon (Falco cf. F. peregrinus) 
Prairie Falcon (Falco cf. F. mexicanus) 



Quail 



(Meleagris 



Quail 



Virginia Rail (cf. Rallus limicola) 



LA-1549 LA-25277 



Pied-billed Grebe (Podilymbus podiceps) 5 (1) 1 (1) 

Double-crested Cormorant (Phalacrocorax auritus) — 7 (2) 

Turkey Vulture (Cathartes aura)*** 1 (1) 2 (1) 

2(1) 

KD 

5(2) 

2(2) 
1 (1) 12 (2) 

1 (1) 1 (1) 

6 (2) 7 (2) 



2(1) 



Canvasback (Aythya cf. A. valisineria) * 3 (1) 

Duck (Aythya sp.) 

Scoter (Melanitta sp.) 

Common Goldeneye (Bucephala t 

Common Merganser (Mergus me\ 

Merganser (cf. Mergus sp.) 

Ruddy Duck (Oxyura jamaicensis) 

Anatidae, indet. 

Bald Eagle (Haliaeetus leucocephah 

Northern Harrier (Circus cyaneus) 

Cooper's Hawk (Accipiter cooperii 

Goshawk (Accipiter gentilis, 

Swainson's Hawk (cf. Butt 

Red-tailed Hawk (Buteo iat 



American Kestrel (Falco sparverius) 2 0) 



2(1) 

1(1) 
3(1) 



2(1) 

KD 
11 (3) 2 (1) 

2(1) 

1(1) 
2(1) 

1(1) 

2 (1) 2 (1) 

1 (1) 1 (1) 

14 (2) 11 (2) 

3 (2) 1 (1) 



KD 
3d) 

KD 

2 (1) 9 (2) 

11(3) 
2(2) 

KD 



94 



EMSLIE, SPETH & WISEMAN Vol. 12, No. 1 



TABLE 1. (continued) 



Taxon 



** 



Owl (Asio sp.) 



Flycatcher (Tyrannus sp.) 
Steller's Jay (cf. Cyanocitta stelleri) 



Sparrow (Zonotrichia sp.) 
Emberizinae, indet. 



Meadowlark (Sturnella sp.) 



LA-1549 LA-25277 



Common Moorhen (Gallinula chloropus) 2 (2) — 

American Coot (Fulica americana) 60 (9) 119 (13) 

Sandhill Crane (Grus canadensis) * 3 (1) — 

American Bittern (Botaurus lentiginosis) 
Great Egret (Casmerodius albus) 



i(i) 
id) 



Dowitcher (Limnodromus sp.) 

Gull (Lams sp.) — 

Mourning Dove (Zenaida macroura) 9 (3) 

Scarlet Macaw (Ara macao) — 

Macaw (Ara sp.) — 
Screech Owl (Otus sp.) 



1(1) 



1(1) 



Northern Flicker (Colaptes auratus) 4 (1) 



1(1) 



Crow or Raven (Corvus sp.) — 

Northern Cardinal (Cardinalis cardinalis) 1 (1) 

Lark Sparrow (Chondestes grammacus) * '* 1 (1) 

2(2) 

1(1) 



Red-winged Blackbird (Agelaius phoeniceus) 2 (1) 



4(2) 



Brewer's Blackbird (Euphagus cyanocephalus) * 1 (1) 

Icterinae, indet. — 

Passeriformes, indet. 

Total 



1(1) 



1(1) 



7(1) 
1(1) 



Great Horned Owl (Bubo virginianus) 5 (2) 2 (1) 

Burrowing Owl (Athene cunicularia) 9 (3) 

Short-eared Owl (Asio flammeus) * 1 (1) 



KD 



Chihuahuan Raven (Corvus cryptoleucus) 6 (1) 4 (2) 



2(1) 



KD 
133 6 

316 (61) 232 (59) 



ISpecies that are migratory and found in southern New Mexico during winter months 
only are indicated by an asterisk (*); spring/summer migrants are indicated by a double 
asterisk (**) (Hubbard 1978). 



sample with their proportion in the sample from Re 
are well represented at Henderson, comprising 



in 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



95 



sample from Rocky Arroyo (14 of 232 bones), but fully 50.0% of the smaller but 
systematically recovered sample from Structure 2 (6 of 12 bones), a value nearly 
identical to the one from Henderson. 




60 



50 



40 



LjJ 

O 30 

a: 

LU 20 

CL- 
IO 



0- 




CO 





on 

o 




CE 




to 




< 




CO 

Ld 




o 




Q. 

< 



15 



9 





CO 




o 





< 





120 



65 



T 



CO 

Ld 




a: 
o 




a: 



HENDERSON SITE 
ROCKY ARROYO 



17 




CO 

Ld 




O 

a: 




CO 



15J 



13 




CO 
Ld 




Qd 

O 




Ld 
CO 

Q- 



FIG. 6.-Percentage of total NISP (see Table 1) per order of bird bones identi- 
fied from the Henderson site and Rocky Arroyo site. The number at the top of 
each bar is the NISP for that order. 



some 



post-depositional 



taphonomic processes, particularly differential bone preservation resulting in 
fewer bones of smaller species at Rocky Arroyo. However, preservation ot 
small and fragile bones was generally excellent at both sites. For example, 
thousands of tinv. delicate fish bones were recovered at both sites, and many 



papery 



Structure 2 fill at Rocky Arroyo. Most 



damag 



damage was most 



those of bison, on which tool marks and fresh breaks were common. The 
bones do not show this damage and display few recent breaks. Most 



96 



EMSLIE. SPETH & WISEMAN Vol. 12, No. 1 



were 



protected them from exposure to weathering and decay. While many of the larger 
bones had become brittle, cracked, and broken, there is little evidence to suggest 
that bones of any size had disintegrated into unrecognizable debris within the 

tins. 

In sum, it seems reasonable to conclude that differences in the avifaunas 

between Henderson and Rocky Arroyo, with the exception of the bias against 

recovery of passerines from lack of screening at the latter site, may have a 



most reliable intersite comparisons will 

which 



much smaller 



Statistical comparisons of the percentage NISP per total NISP by orders 
represented at each site indicate that significantly more bones of Anseriformes 
and Gruiformes, especially American coot (Fulica americana), were recovered 
from Rocky Arroyo (t s >3.6, p<0.001), and more of Strigiformes and Passeri- 
formes (the latter, as noted above, probably reflecting recovery bias) from 
Henderson (t s >4.9, p<0.001) (Fig. 6). In addition, bones of macaw and cor- 
morant (one with cut-marks) occurred only at Rocky Arroyo, and bones of 
mourning dove (Columbiformes) were recovered only from the Henderson site. 
The absence of this last taxon from Rocky Arroyo, however, also may be due 
to recovery biases of small bones at this site. These comparisons suggest that 



more 



Henderson 



muskrat 



Arroyo in comparison to Henderson (Wiseman 1985; Rocek and Speth 1986). 

Intrasite comparisons.— The lack of secure provenience information for most of 
the avifaunal remains from Rocky Arroyo precludes intrasite comparison of the 
distribution of bird bones by species or body part. The sample from the Hender- 
son site, however, can be analyzed in this manner with comparisons of bird 
remains from the room blocks with those from the midden associated with the 
roasting feature in the east plaza. Sample sizes are too small to permit comparisons 
among individual rooms. 

These comparisons, best expressed in terms of the number of bird bones per 
cubic meter (m3) of deposit, indicate relatively similar and low bone densities 
in most of the room blocks and in the east plaza (Table 2). In contrast to these 
modest values, the trash deposit in room C-5, the only significant midden deposit 
found in the rooms sampled to date, produced a density of bird bones (11.631 mP) 
that is nearly 2.5 times greater than the east plaza value. The fill from room C-5 
included more than half (53.4%) of the passerine remains recovered from the site, 
and their density in the trash (8.10/m3; see Table 3) of this room was nearly 
7.5 times greater than their density in the east plaza trash (1.09/m3). Differences 



few NISP 



rooms compared 



Avifaunal exploitation. —Water birds are important in pueblo symbolism in rela- 
tion to annual rainfall and the onset of the growing season (Tyler 1979). Conse- 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 97 



TABLE 2.— Density (number of specimens per cubic meter of deposit) of avian 
bones by provenience at the Henderson site (all taxa combined). 



^^— 



Total Avian 
Total Volume Total Avian Density 

Provenience Excavated (m 3 ) NISP (bones/m 3 ) 



Center Bar 
All Rooms 



25.76 166 6.44 



Room C-5 only 8.77 102 11.63 

Excluding Room C-5 16.99 64 3.77 



East Bar 



34.85 83 2.38 



East Plaza (Trench C) 11.90 57 4.79 



TABLE 3.— Number of bones (NISP) and density of bones per cubic meter of 
deposit (in parentheses) for principal taxa (order) by provenience at the Hender- 
son site. 



Order Center Bar Room C-5 East Bar East Plaza 



Gruiformes 34(1.32) 1(0.11) 29(0.83) 11(0.92) 

Accipitriformes 11 (0.43) 5 (0.57) 3 (0.09) 6 (0.50) 

Galliformes 11 (0.43) 7 (0.80) (0.00) 2 (0.17) 

Strigiformes 5(0.19) 6(0.68) 2(0.06) 4(0.34) 



>rmes 
irmes 



108 (4.19) 71 (8.10) 40 (1.15) 13 (1-09) 

11(0.43) 6(0.68) 5(0.14) 5(0.42) 



quently, duck and goose wing fans are used in ceremonies rela 
agricultural cycle. Moreover, certain of these species are migratory 
only in the winter in southern New Mexico (Table 1). These species we 
primarily during late fall through early spring (Hubbard 1978). Bones i 
red-tailed hawk, quail, and raven suggest a late spring and summer e 
of these species. 

Many passerine species are also symbols of rain for Southwest 



Ame 



ct, it is interesting that a significantly greater percentage c 
elements comoared to lee elements were recovered from 



site (Table 4), 



porta 



Wing bones also outnumber leg bones at Rocky Arroyo, but tne nuim* 
are too small for reliable comparisons. Although most of the passenne bones rro 
Henderson could not be identified, those that were include species common 



98 



WISEMAN Vol. 12, No. 1 



TABLE 4.— Comparison of the distribution of wing (radius, ulna, carpometa 



emur, tibiotarsus, tarsometatarsus) elements 



Henderson 



Henderson site Rocky Arroyo 

Order NISP % NISP % 



Anseriformes: wing 5 55.6 12 70.6 

leg 4 44.4 5 29.4 

Accipitriformes: wing 8 44.4 8 38.1 

leg 10 55.6 13 61.9 

Galliformes: wing 4 50.0 1 12.5 

leg 4 50.0 7 77.5 

Gruiformes: wing 13 34.2* 26 34.2 

leg 25 65.8 50 65.8 

Passeriformes: wing 65 63.7* 6 66.7 

leg 37 36.3 3 33.3 

Total Aquatic* w i ng 21 39.6* 39 39.4 

leg 32 60.4 60 60.6 



1 The category for total aquatic taxa includes bones of Podicipediformes, Pelecaniformes 
Ciconiiformes (except Turkey Vulture) and Charadriiformes in the NISP. An asterisk (* 
indicates those proportions that are significantly different (t s >2.1, p<0.05). 

associated with agricultural fields, such as icterids, corvids, and sparrows (Emslie 

1981a, 1981b, 1983). 

While wing elements outnumber leg elements in passerines, this is not so 
in the other orders that are represented at the sites (Table 4). In the aquatic 
birds, especially Gruiformes, leg elements significantly outnumber wing ele- 
ments. While this in no way precludes the use of aquatic bird feathers for ritual, 
fletching arrows, or other purposes, the abundance of leg elements suggests that 
these birds were used for food since the legs contain greater muscle mass than 
the wings. However, only five bird bones (less than 1%) from the two sites were 
burned, and most or all of these derive from species that were probably of little 
or no importance as sources of food: a humerus and an ulna of unidentified 
passerines from Henderson, and two Corvus sp. tibiotarsi and a tarsometatarsus 
of Falco columbarius from Rockv Arrovo. Moreover, none of the bones of aquatic 



burned 



were 



elements sueeests that roasting was not a common m 



of preparation. This contrasts to bison: about 6% of the bison bones are burned. 
Interestingly, turkey (Meleagris gallopavo) , probably an important source for food 
as well as feathers in the late prehistoric Southwest, is rare at both sites; only 
two bones were recovered from Henderson and nine from Rocky Arroyo. The 
apparent scarcity or absence of captive or domestic turkeys at the Henderson site, 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 99 



com 



from eggs large enough 
fragments do derive from 



domestic 



from 



number 



aquatic birds at both sites, suggests specialization in the use of avian resources 
by the occupants of these sites. Avifaunal specialization by Pueblo Indians is not 
unknown in the prehistoric Southwest. Over 2,000 bones of turkey (Meleagris 
gallopavo) from Sapawe Pueblo (A.D. 1400-1550) and Pottery Mound (A.D. 1325- 
1490; 88.4% and 64.0% of total avian NISP from these sites, respectively), and 
hundreds of bones of golden eagle from Picuris Pueblo (A.D. 1250-present; 34.4% 
of NISP) and sandhill crane from Pottery Mound (10.0% of NISP), indicate large- 
scale exploitation of these taxa specific to those pueblos (Emslie 1981a). At Rocky 
Arroyo and Henderson, the large number of coot bones is unusual. This species 
was poorly represented (<1% of total avian NISP) in the large samples of bone 
from Pottery Mound, Picuris, and Sapawe (Emslie 1981a). We speculate that coot 
may have had an important symbolic function at Rocky Arroyo and Henderson, 
similar to that of other aquatic species. 

Site-specific avifaunal specialization in the Southwest may have been encour- 
aged by the development of trade centers for distributing feather blankets, food, 
and whole birds or feathers for ceremonial use in puebloan religion. The sym- 
bolic role of birds is apparent from preserved kiva murals at Kuaua and Pottery 
Mound (Dutton 1963: Hibben 1975), and through ethnographic studies (Tyler 



kinship 



>n also may have developed in relation witn parti 
ups that occupied specific pueblos (Emslie 1981a) 
from Henderson and Rocky Arroyo, it is possible 



communities were gathering coots and passerines 



ehistoric trade of birds. —The scarlet macaw bones at Rocky Arroyo indicate 
ide with Mesoamerica. At least one individual is represented by eight bones; 
eluded are elements of the skull, wings, and legs. The remains may have 
en part a single macaw burial, or one or more individuals from diverse proven- 
ices. In addition to the macaw at Rocky Arroyo, seven copper bells (another 
*de item from Mesoamerica) were recovered from Bloom Mound, a roughly 
ntemporaneous pueblo located within 2 km of Henderson (Kelley 1984). 

Macaw bones and feathers are known from prehistoric sites throughout the 
>uthwest as far north as southeastern Utah (Hargrave 1979). The bones from 



hist 



easternmost 



these bones represent one of the latest records of macaw 



Mesoamerica 



and Kelley 1975). This breakdown is supported by negative evidence at Pottery 
Mound, a pueblo established about A.D. 1325, or approximately when Rocky 
Arroyo was abandoned (though the age of this site may be more correctly placed 
at post-A.D. 1400. E. Charles Adams, personal communication to Speth, 1991). 



100 



EMSLIE, SPETH & WISEMAN Vol. 12, No. 1 



This large pueblo produced thousands of bird bones, but none of macaw despite 
its frequent representation in the many kiva murals preserved there (Hibben 1975). 
A revival of macaw trade may have occurred in the 1400s (Hargrave 1970). 
Apparently, Rocky Arroyo represents one of the latest pueblo sites in the 
Southwest to be involved with trade from Mesoamerica before the breakdown. 
Another possible trade item with Mesoamerica is the northern cardinal. 
This species was rare in Arizona and apparently absent in New Mexico prior 
to the 1870s, but expanded its range northward and eastward in the following 
decades (Phillips 1968). It did not become established in New Mexico until the 
early part of this century. We know of no other archaeological records of cardinal 
in New Mexico, though there are three records from Arizona with the earliest 
dating from A.D. 850-950 (Ferg and Rea 1983). These authors suggest that 
cardinals, because of their bright plumage similar to macaws, may have been 
brought into the American Southwest from Mesoamerica through trade. This 
suggestion assumes that the species only recently expanded its range northward, 
and did not formerly occur in the Southwest where it may have disappeared for 
unknown reasons prior to historic times. Until this problem is resolved, it is 
possible that the specimen from Rocky Arroyo represents an additional trade item 
at this site. 

CONCLUSIONS 



Excavations of the Rocky Arroyo and Henderson sites have provided diverse 
avifaunas that are the first of their age to be reported in detail from southeastern 
New Mexico. The presence of the scarlet macaw (Ara macao) at Rocky Arroyo is 
the easternmost record of this species in the prehistoric Southwest and, in 
conjunction with the copper bells from nearby Bloom Mound, and possibly also 
the cardinal from Henderson, indicates thirteenth and/or fourteenth century trade 
links between the Pecos Valley and regions to the west and south. Moreover, 
we believe these sites present additional evidence that specialized exploitation 
of avian resources, perhaps for trade, was occurring in the late prehistoric 
Southwest. This specialization probably developed in conjunction with an 
agricultural economy; many species of birds are highly symbolic among modern 
Pueblo Indians, especially in relation to the growing season. 

The preponderance of aquatic species at Rocky Arroyo may reflect speciali- 
zation at this site for birds that were used for food and religious/ceremonial 
purposes. A similar specialization in aquatic species was not apparent at the 
Henderson site, though it is located on the same drainage and is roughly con- 
temporaneous with Rocky Arroyo. Henderson occupants may have specialized 
on the acquisition of passerines instead of waterfowl, but excavation biases 
may have caused an unbalanced recovery of these small birds from these sites. 



NOTES 



lit should be pointed out that the name Rocky Arroyo also has been used for a late 
Pleistocene cave fauna and archaeological site from which fossil bird bones were reported 
by Wetmore (1931, 1932). However, this cave is located in Eddy County, northwest 
of Carlsbad. New Mexico, and is not the same Rockv Arrovo as discussed here. 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



101 



2Proveniences associated with lot numbers 1-2027 have been published in Rocek and 



342) 



mens from flotation samples, have not been published but are available from JDS. 

3A11 Rocky Arroyo specimens are deposited at the Laboratory of Anthropology of the 
Museum of New Mexico in Santa Fe: the Henderson matprialc aro mrrontiv «« i™„. 



term 



Michigan 



Conservancy 



The Henderson site has since been donated to The 



ACKNOWLEDGEMENTS 



We acknowledge the generosity of Matthew and Karen Henderson, who made the 
University of Michigan excavations at the Henderson site possible. Jay Hollifield granted 
permission to the Museum of New Mexico to salvage the remaining fill from Structure 2 
at Rocky Arroyo. Financial support for the analysis of the avifaunal remains was provided 
by the Museum of Anthropology of the University of Michigan. Mapping of Rocky 
Arroyo was done by William F. Turney. Graphics were done by Kay Clahassey. We also 
thank Amadeo Rea for helpful comments and suggestions in preparing this manuscript. 



LITERATURE CITED 



DUTTON 
Way. 



Albuquerque. 
SLIE, STEVEN 



The 



305 



. 1981b. Prehistoric agricultural 
ecosystems: Avifauna from Pottery 
Mound, New Mexico. American Anti- 
quity 46:853-861. 

1983. Cultural and climatic 
implications in Anasazi faunal exploita- 
tion: A review and perspectives. Pp. 119- 
123 in Proceedings of the Anasazi Sym- 
posium 1981. Jack E. Smith (editor). 
Mesa Verde National Park and Museum 
Association, Colorado. 
FERG, ALAN and AMADEO M. REA. 1983. 
Prehistoric bird bone from the Big Ditch 
Site, Arizona. Journal of Ethnobiology 
3:99-108. 

HARGRAVE, LYNDON L. 1970 Mexican 

MaCaWS. UnivprciHr r.' 



Arizona 



Pological Paper No. 20. 

. 1979. A macaw feather arti- 
fact from southeastern Utah. South- 
western Lore 45(4): 1-6. 
ttlBBEN, FRANK C. 1975. Kiva Art of the 
Anasazi at Pottery Mound. KC Publica- 
tions, Las Vegas. 



HOUGHTON, FRANK E. 1974. The climate 
of New Mexico. Pp. 794-810 in Climates 
of the States, Vol. 2: Western States. 
United States National Oceanic and 
Atmospheric Administration, Water 



Washingt 



New York. 



HUBBARD, JOHN P. 1978. Revised check 
list of the birds of New Mexico. New 
Mexico Ornithological Society, Publi- 
cation No. 6. 

KELLEY, J. CHARLES and ELLEN ABBOTT 
KELLEY. 1975. An alternative hypothesis 
for the explanation of Anasazi culture 
history. Pp. 178-223 in Collected Papers 
in Honor of Florence Hawley Ellis. Theo- 
dore R. Frisbie (editor). Papers of the 
Archaeological Society of New Mexico 

No. 2. 
KELLEY, JANE H. 1984. The archaeology of 
the Sierra Blanca Region of southeastern 
New Mexico. University of Michigan, 



Anthropology, 



ical Paper No. 74. 



CHARMION 



\Q£ 



• It 



South- 



west Indian Turkeys. Southwest Bird 



Arizona 



OLSEN, SANDRA L. 1979. A study of bone 



AZ 



The 



102 



WISEMAN 



Vol. 12, No. 1 



LITERATURE CITED (continued) 



PHILLIPS, ALLAN R. 1968. The instability 
of the distribution of land birds in the 
Southwest. Papers of the Archaeological 
Society of New Mexico 1:129-162. 

ROCEK, THOMAS R. and JOHN D. SPETH. 
1986. The Henderson site burials: 
Glimpses of a late prehistoric popula- 
tion in the Pecos Valley. University of 
Michigan, Museum of Anthropology, 
Technical Report No. 18. 

SOKAL, ROBERT R. and F. JAMES ROHLF. 
1969. Biometry. W.H. Freeman, San 

Francisco . 
TYLER, HAMILTON A. 1979. Pueblo Birds 

and Myths. University of Oklahoma 

Press, Norman. 
VON ESCHEN, G.F. 1961. The climate of 



■ 

New Mexico. University of New Mexico, 
Bureau of Business Research, Business 
Information Series 37. 
WETMORE, ALEXANDER. 1931. The Cali- 
fornia condor in New Mexico. Condor 



77 



. 1932. Additional records of 

birds from cavern deposits in New Mex- 



34 



WISEMAN 



and sedentary occupations: Startling 
data from Rocky Arroyo (LA 25277), 
Chaves County, New Mexico. Pp. 30-32 
in Views of the Jornada Mogollon. 
Colleen M. Beck (editor). Eastern New 
Mexico University, Contributions in 
Anthropology 12. 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



103 






Q 

2 
3 

2 

» • 




3 

o 



Oh 



u 

en 

CD 

Q 



cd 



o 

X 






03 
C 



cn 

no 

CM 




en 
3 

01 

Q 



CO CN 




N V 




r-4 c/3 




rn 3 




v en 




en ^j 




p 03 




Oh O 




>H -^H 




03 X 


CN 


u *jrt 




as "^ 


• 


QJ hJ 


O 


C ^ 


J 






LO £ 

cm £ 



03 

3 








en 

.y 00 



en 
CD 

£ 

O 



en 

3 




73 

cd 

5, 



u 



O 

Oh 



CD 

O 
ft. 



en 

cd 

03 

5 



hJ 



03 

£ 

en 
en 

03 

CN 

i 



03 
3 



o 

£ 

o 

9h 



en 

£ 

CD 

6 

3 
X 



x 

cd 

Q 

o 

en 

03 

£ 

i 



3 

3 

a" 

a 03 

O en 



J 

&H 



O 
en 

cd 

£ 
3 

CN 

m 

CN 

I 

< 



en 

2 

CD 

£ j 



O 
-J 

en 

en 
03 



03 



en 

03 



en 

CD 03 

X o 

Q s 



CD 

s 

o 

en 

03 



■J 

Q 

CN 

O 



en 
3 




CD 
03 

•a 

03 

J- 

O 

o 
ca 



I 

o 



CO 

00 







ft- 



c 

o 

en 
en 

CD 



ON 

03" 

c 

Q 



CN 

o 

en 

j> 





ex 



en 

CD 

£ 



CD 

03 

T3 



3 

en 



3 



en 

CD 
03 

2 

8 



c3 J 




o 



P 
> 



£ CD 






3 



CO 



3 

be 






*CD 

4^ 



en 

CD 
03 

en 

C 

CD 

s 



en 



03 

03 
U 



c 

03 

CD 
> 

O 

o 



On 
IN 

ON 




60 

o 
o 

CD 



S 



CD 
CD 






u 



en 

O 

C 
03 



03 

en 

6 

o 



c 

CD 

en 

g 

CD 

3 



3 



C 
u 

CD 



CM 

1 



o 

3 

0) 



3 

Q 




O 






CN 
CO 

00 



u 

u 

3 



U 




en 
Si 



o 

en 

V-i 

03 



CM 



u 




en 

S 

CD 

£ 

CN 
LC 
CN 

< 



en 

eu 

s 



00 

LO 
IN 




s 




U 



X 



T3 

03 

CD 
»-. 

Q 



en 



C 

CD 

£ 

CD 

3 
en 

03 

CD 

£ 

cn 

2 

CD 

£ 

DC 



1 
CD 

£ 

o 

u ., 

d £ 

£ E 

£ 

°o in 



CO NO 

CM ^ 




vO 



q x 

06 "" 

(N 



X 



to 

c 

CD 



4 j 

C c 



03 



03 



CD 

C 

03 



en 
3 

03 

u 

OS 






13 



o 

03 

O 
C 

CD 

"c3 
U 

CD 

£ 



en 
3 

CD 

£ 

3 

x 

Q 



x 

c 

x 

c 



CN 

1 



r % 




o 

3 
a, 

03 
U 

en 



en 

3 

H 

03 
U 
03 




*>3 

5 N 



en 
en 




00 
IN 





s 

es 






OQ 



en 
3 

CD 

03 

c 
^3 



104 



EMSLIE, SPETH & WISEMAN 



Vol. 12, No. 1 



T3 

P 



o 




Q 

pj 

p< 
pj 



Q 

pp 

Ph 
Ph 

< 



c 
o 



a> 

Q 



c 

a» 



c 
o 

H 



o 

a> 

c 

03 

.a 

CO 



O 



u 




X 



*-• cd 

73 

g 

5 



X 
CO CD 

o 



CD 

C 

bJD 

O 



CD 

CD 



03 



£ 



OS 



en 



03 
> 

O 

o 



CO 

c 

CD 

£ 



u 

CD 

JX 

CD 



73 

cs 

cj 



CD 



U 

CD 

Dm 

CO 

CO 



c 

O 

£ 

S w 



c 

a; 

6 

CD 



o 



C 

60 

O 



£ 



CO 



O 

03 

'C 

3 

C 

£ 




C 

03 

O 



C 



c 

•p 

a; 

CO 

C 
< 



CO 
CO 



6 

CO 

03 



73 

C 

c 




CO 
CO 



£ 

A3 

C 



.« 



6 

3 

<N 

73 



O 

03 

O 



(N 
i 



CO 

S 

£ 

9 

D 

O 

-J 



CO 

3 

CO 

03 



73 

C 

CD 

60 

c 



CO 
CO 



£ 

CO 

3 

03 
U 
03 



03 





£ 



CD 

£ . 

O CO 



CO 

o 




u 

R 




00 

in 
in 





-3 

CO 



CO 

3 

CD 

03 

C 

c 




73 

C 

03 




73 



O 

03 

O 

u 



0) 
03 

IN 
IN 

<N 

in 

<N 
i 

£ 

o 



o 



03 




CX 



X 



3 

(U 

£ 

a» = 

CD 

< o 



eft 

.a 

CO 

I 
CD 



03 

£ 

o 



73 



O 

u 

03 

O 



73 

hJ 03 

.k CO 




CN 



CO 

3 




m 

73 



cy 

03 




c 

03 

O 



rsi 

m 

00 

v ON 

03 t^ 

C 



3 






CO 

S 

£ 

3 



I 

c 



03 

£ 

CL> 

o 



0) 




52 

u 

CO 




O 

5 



u 



a; 



03 



03 

c 

3 

73 

03 

£ 

o 



3 



C 

CD 

73 



C3 
. CO 



SX 




o 



O 



CN 

in 




NrC ?H 



CO 

co 



cu 

03 

CO 

a» 

C 

o 

6JD 
C 






03 



£ 

OS 

Ih 



O 



^ ft 

CO 



c 





03 

u 



03 



CO 

3 



•J 



CM 
I 




T3 

£ 

s- 

3 



^ CO 

T3 73 



Q 

73 

C 

CD 



CU rsi 



co" 

2 

CD 

£ 

C 

fl5 



Pi 

D 

£ 

c 

CD 



CO 



O 





CM 
i 



O 
-J 

03 

c 




3 






CO 






^ 



Tt< 



00 







O 
co 



CO 









Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



105 



U 



o 




Si 

05 

ex 

CD 

c 

o 



s £ 



05 

CD 



CD 

03 

CD 

3 

3 

X 



'3? 

CD 

o 

a, 

T3 



CD 

u 

05 



O 



c 






e g 



05 

CD 



C 



c 

c 



05 
CD 



3 
X» 

<d 

CD 

03 




c 

X. 

CD 



CN 

i 



o 

CD 

CX 

03 rV 

03 



6 



cd 



CD 
3 

Ch Q 




V * 



IT) IT) 

< 



£ 

C 

5- 

0) 



CD 



o 

£ 
3 



3 
C 

£ 



CN 
i 



(N 



o o 



ON 



o 

u 

05 

o 

. £ 

as g 

lo ^ 



- fr 1 

5 fc 



CN 

O 

CD^ 

3 
CX 

05 



s ° 

05 v 

CJ CD 



£ .2 

CN S 

LO 
< ^ 



T3 

<d 
CX 

60 

C 



CD 

CD 



£ 

CD 
3 

£ 

3 

PCS 



i 



l 

05 

O on 

u 




ON 
LO 




& ° 3 



O 

u 

05 

J- 

O 

OS 

a; 
*os 






03 



LO 



3 ON 
&LO 

05 
U 

CD 






K on 

ON * 

T^ CD 
LO ~0 



CD 

LO 

00 



• v 03 



i 
05 



CO o 



CX 



I 

05 




i 





o 




w 

LO 

nO 

LO 

2 

LO 
CN 

LO 

05 S 

c 
Q 



P4 CM 



^ D 



6 

o 

CD 

OS 



• V 




CO 



CD 

3 

CD 

03 






03 00 



CD 

3 
03 52 



cd 

£ 



CN 



03 



05 



c d 

O * 

C co 

05 LO 

U 00 



CD 

3 

CD 

05 



05 



CO 




6 

c 

CD 





CD 

3 
CD 
U 




O 

CD V 

3 

ai 

£ 

3 

OS 




CN 

O 
3 

6 



hJ 



(N 

C 

CD* 

3 

£ 

3 

D 




O 




03 




U 




5 




a) 

6 


CN 


o 


* 


!- 


O 


u 






- 


hJ 


3 


D 


£ 


• • 


a> 


1% 


v*- 


CN 


a; 


LO 


» « 


CN 

i 


i 




CD 

3 
CX 

05 
U 

05 



E 

CX 
03 

PS 



rsi 




C 



CX 

CD 



« 

^ 

« 



Q 






00 

R 










CD 

3 

05 

C 



hJ 



52 °0 

£0 



g CD 

* 3 

CD d> 

3 05 

5 j 



CX 

CD 
CD 

PC 






u 



CD 



CD 

^ ON 

y oo 







CD 

c 



c 



<3 U 



03 

T3 



05 

c 

< 



CD 





51 



NO 




CD 

3 

05 
C 

c 



hJ 



NO 

cd L\ 

3 r-i 

li 

w C 



U^ 






o 



<u 



00 

CNJ 

00 




00 

LO 



CD 
3 

CD 
03 

C 

c 



SX 



y 

^ 



t; -2 

03 r^ 



106 



EMSLIE, SPETH & WISEMAN 



Vol. 12, No. 1 






O 





D 

3 

Ph 
W 

s 

x 



Oh 



c 

o 



Oh 



G 

CD 

CD 

Q 



c 

£ 

CD 

w 



c 

o 

H 



i 

03 



CD 

6 

o 

CD 

OS 



CD 

3 

CD 
OS 



i 




Q 



o 

IS 



<N 

o? 

C 

j en 
Q N 



• ■ 
ON 

lo 



Q 

3 

CD 



00 

ON 
ON 

w 

On 

s 

ON 




l 

3 



CD 

3 

CD 

03 



• - 

IN. 

in 

rsj 

in 

<N 

i 

< 

-J 



ID 

3 



CN 

c 



c 



03 



6 



CD 

03 

43 



CD 

3 
CD 

Q 



CD 



£ 

o 



5 K 



o 



CD 



CD 
C 

CD 

£ 



<3 

B 



03 

O 

CD V 

3 
O 

o 

a, 



o 



CD 

CD 



60 

3 



< 



CD 

N 

CD 



LO 



to 

CD 



CD 

3 



-J 

Q 



y 

o 

CD" 

3 

CD 

o: 



T3 

03 

Q 

NO 

CO 

CO 



T3 



C 

jS 



CD 

o 

ON . 

T^ O 

LO IN 



K 



c 

u 

03 

o 

* 9 



ID 

on 

Dh 

LO 

CO 

ON 
CD V 

3 



CD 

3 

a, 

S £ 

CD cD 

03 
U 



LO 

CN 

CO 



6 






CD 



CN 

Dh CD 

03 



03 



3 

CD 




I-J 

CN 
LO 




LO 



Oh 



CD 

o 

CD 

03 



ON 





OS CD 
NO S 

CM X 



5- 

3 

£ 

CD 



i-J 

CO 

00 
CN 



9h 

3 
CD 

w 

co 

CO 

LO 

CO 



CO 

CO 

co Q 



LO co 



CD 

3 

CD 

03 



o 

IS 



LO 



ON 

CO 

CD V 

3 

CD 

1- 

03 



ON 

LO 

CN 



NO 



O 3 

J £ 

CD 



CD 

3 

J5 



hJ 



en 

£ 

3 CU 



CD 
03 

3 



CN 

i 



i 

O 

CD 

03 





O 

CD* 

3 

CD 

03 



Oh 
CN 

O 

03" 

O 



CD 
OS 

3 
Q 

IN 

On 
On 

CO J2 

3 

ex 

03 

CD 



s £ 



& 



CD 



o 

CO 

3 

CD 

- 

OS 



s 



CN 



03 



CD 

£ 



tt5 

Oh 

ON 
LO 



LO 







00 

ON 

CO 



CM 

o 

cd" 

3 

03 
03 



CD 

£ 

o 

& 

03 
U 




oo 

00 



3 



o 

CD 



I 

CD 



00 
CO 

oo 



CD 

£ 

CD 



8 



3 



CD 

t 

03 

OS 

3 

O 
CO 



CD 

8 



2 lo 

^j LN 

CD 



C3 



Dh 

CD 

O 



^ 




CQ 



QQ 



P 

5 § 



cj 



^C 



I 

03 



CD 

£ 

o 

CD 

OS 



D 

LO 

LO 

00 
03 V 

Ph Cn| 

CD 

3 

CD 

H 
OS 





CD 
03 

T3 



3 
O 



CD 
CD 

o 

2 



00 
LO 
IN 



CD 

3 

CD 
OS 

3 

3 



^J 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



107 



T3 

c 

3 

CQ 



O 



CD 



• 



2 s 

£ c 

o •— 

00 e 
,3 CD J5 

y 




CD 

CD 

6 

o 



-. -rf H3 





CD 

0) 

QJ U > 

A 

NO 

00 

ON 



c 




6 

CD 



CD 




.a o o u 




55 CD "S5 CD 

3 73 .2 cd 

£ u X g 

u cd cd g 

^ X fXTJ 



a» 

CD 

<d o 
a; 

•a as c 

CD 03 r- 
P ^ S 

c o * 



« £ s 

2 rx 

? a © 



>H CD 2 



c 



c 

<d 

CD 

CD 

s- 

cx 

CD 
CD 



C 

CD 

E 

a; 



cd 



cd 



g 

ex «> 

£ S3 

o 




CD CD 
CD C 

O QJ 



>-» 



I 




I 



a 

S 



c 



CD 



ON 

CN 




> U 

3 O 



£ 



CD 

P 

a! T3 

5 -a 

^ QJ 



03 73 

CD 



K 



cd u 
Si CL, 

t||<3 



c 

03 



cd 

CD 
CD 



•SB * 

Oc ^ x. 

Up— - \J 

3 TJ £ * 



cd 

CD 




<D 

3 



3 >^ C 



5 
X 



<D 

CD 

O 



cx § 



a* 

CD 






S (A 

& o 



<5 



5 



si 

CD ,^> 

<3 



o 



03 



C 

at 

£ 

03 

v- 



C2 

sx 



w 



■1 

C 



03 



03 



CD 



T3 



•^ 








a/ 






03 



w ^^^"* 

X 



C 






o fc 



-c 



03 
— — . vj 

r- ^J fZ 

03 



OJ 



i 

E 

c 

i 

^3 



1> 




o 



sr 

c 



C 2 

OJ ° cv 

H <0 d) 

^ s ^ 

c 5 CD 

OJ •" 



o 

CJ 
03 

03 t 

CD O 

CD 



O 



£ 



CD 
OJ 

CD OJ Jj 

O Clq, 
D-LD cd O 



Oj C 

C 

u 

c 

03 




CN 

O 

CD" 

3 

CD 

03 



a) 

£ 

o 

CD 

03 



Q 

IN. 

CN 

LO 
(N 




Q 



tF 



• o . 

3 

£ 

CD 



C4 

LO 

CN 
i 

< 



CD C CD 

3 p P 

C ,oj C 

3 .. 3 
Q 



• . O • • 

i 



3 < 



Q 

CO 
CN 
00 

CD V 

P 

CX 

03 

03 



OJ 

03 « 

B5 

CX 



On 

LO 



CD 

P 
CD 

03 



CD 



3 



P 

£ 

03 03 <+h 



P CD 

CX g 



u 

CD 



-- 1 rv^ 



CD 
P 



o 

CD 

i- 



o 






Pi 

(X 

(N 

in 



o 



CD 



CD 

H 

03 



O 



<N ^5 

2" 



p 

£ 

CD 



CX co 

cn y 



CD 

P 

CD 

03 



^ 



2 85 



— 

OJ 



O 

c 

CD 

Q 

•P 03 
CD -*-» 

fe 



CD 

£ 



73 

C 

CD 

^ CX 
oj ^ 



£ 



ON 

P ON 



ON 



X 



CO -- 

p 



03 



12 * 

8 « 

03 ^ 

o 




3 

'5 

03 

O 



LO 




Q _ 

.P -N 

.2 ^ <N 

v ON 

CD 

u 15 



£ 

CD 

P 

J- 

CD 



Q 

CN 

CO 

CD V 
P 

u 

CD 

g^> cj 
3-C £ 



I tu .2* 





S 






- 3 p 

P X CD 



m 03 



03 

C 



3 O 

£ E 

CD 03 



nC 
(N 

00 

NO 

CO 

LO 

CD 

£ 

(X 

CM 

LO 



• 

5 



S 






00 
cd \n 

C 

» CD 
P 

CD 
03 



CD 






c 



tin J 



„ 3 



S 



^"giC 



CD 

S^ LO 

C3 00 

G 



CD 

00 
CD 






CD 
OJ 

£ 



CD 
03 

73 



C 
03 



03 



a. 00 

5 LO 

C5 



CD 



oc 



00 „ 

CD 

P 
CD 

2 

CD « 3 

03 



CD 

C 



OJ 

03 

C 

•S 

o 

CX 

o 



c 

c 

O 



g 
5 



<3 



sr 




v oc 



s 

00 



a> 



108 



EMSLIE, SPETH & WISEMAN 



Vol. 12, No. 1 



to 



U 

03 



~ 




C 

o 



Oh 

■c 

Q 

CO 

D 



U 

o 






co 

<3 



c 

E 

W 






cn 

O 
co 

03 



03 



O 

CO 

i- 



ps 




<3 



o 



CO 

IT) 

CO 






p 



tN 



C 



IT) 



i 
< 



o 

05 



*h 

JO 

3 

c 

OS 

£ 

CN 



<N 

ON 

c 

s 




B 



On O 



co 

s^oo 
p in 
in 




oo 




CO 



c 

o 

03 

H 



CO 

s 

s 



oj 



o 




u 



3 

13 



0) 



CO 

3 

0) 
OS 

C 

5 



2 

s 



ss 






hJ 



3 






03 

05 



in 

00 





00 



I 

03 

O 

u 



2 CO* 

03 




0> 




CO 

hJ 

* V 

IN 

m 

CO 



O co 

y 73 



03 

o 



c 

a; 



co 



00 
CN 

as 



3 
C 

>H 

ft « 



03 

Vh 

CO X 

3 m 

Oh 
03 



CO 

C 

3 

if 

C 

03 



tN 



CN 



IN 





in 



CN 
CN 




3 
co 



03 



03 

-J 

Q 



CO 
CO 



s 



co CU 




CO (NJ 



ON 



CO CO -^ 

73 XJ vo 



co CU 




O O ^ 



w in 

CN CN 



3 

CO 

W 

00 
CN 

CO 




O S 7 

Oh tf Ph 

a> 

s 

o 

(X 

03 cN 



-J 




in 



in 

co 




03 

Vh 

O 

a; 



tf 



CO 

H 

OS 




O 



CO 



CO _Q 

3 



CO 

3 ^ 

CO iS. 

!3 ** 




00 j 

■"a 







m 



rH IN 

..3 w 



co v rM 

it 



o 



co 

CO 



O 00 



i 

O 
co 

3 jS 

S3 « 



Oj «N 



1) 



^ 




i£2 £ 




03 

C 
3 




m co Pi 



SO ON 




CN CN 



IN 
CN 



CO 

If) N LO 

Oh in 
S co 



05 

c 

Q 



CV 

£ - 

Oh in 

P5 3 

Q £ 

cy 



S^ co 

co 3 

CO 

H 
03 



v^) CO 

o 

oo ^o 

CO 



o 

CO 

03 



5 

CO 











m 

rH 




CN SO 

o oo 

ON 00 



00 ^ 

oo .5 



ON 




ON 

oo 

IN 



c 



a» 

6 

(J 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



109 



4f< 



in 




Tt< 



ON CD 

^ 13 



CD 

73 



03 



Hi 



o 

CD 

60 03 
en 



6 



^ 03 
TO Q 



Q 



cod 






is. 



03 

O 

£ 

0) 



CD 

CD 



iP in 



i 

< 



la 

03 5 

U 3 










<3J 



CD 

73 



O 

u 

03 

o 



73 

^ v £ 

p 03 



CO 



*. CD 
CD 3 



co .fa 



03 

C 

CD ^ C 
CD •!-» «3 






03 
C 



£ u ^ 



6 i£ 

hJ pL, TO 




13 

C 

C 




TO 



2 ^ 



6 

o 

CD 

os CX 



CD 

OS J 



CD 

03 



o 




-3 

D 




3 

D 




03 
£ 73 

CX, 03 



CO 

73 

c 

OS 

D 

CO 



i 

03 



£ 

o 

CD 

03 



03 ^ a» 



C^ 



73 

g 73 

£h 03 



CD 
CD 



SCD 
03 



CD O 

03 ^D 



O 
3 Pi 



vX) 



»N 



03 C 

a; 



03 



£ 

o 

CD 

03 



Q 



(N 

CD 

3 

CD 

03 



bo £ 

£ * 

35 o 

CD ""^ 

73 

g 

03 



£ 



e 



CD 

03 



03 



£ 

o 

CD 



CD 
O 







CD 

= i 

C 



CD 



CD 



P 

c 

03 

£ 

in 



i 



Q 



i 

fX £ 

5 ° 




u 







a- 
13 « 






" O 



oc 




03 

y 

CD 



c 

03 




° 05 "T c 



03 ^ ^ 



c 

03 



1?^ 



o 

u 

03 H 



£ 




« c 



03 
U 
03 



CD 

£ 




J" 



CD 
►. CD 

Cu o 








a> 



E 

c 

CD 
>- 

03 



w cJ 
" O 



£^ 



« 



CD 

3 

CD 

•*- 
03 



13 

C 



03 



73 ie c^ 

p +* 61 

c 

3 Q 

03 



0» 

£ 

o 

c ^ 



03 03 



CD 

3 

CD 

03 



03 



£ 

o 

CD 

03 



CD J Di 

£-Q 

CD 

3 

CD 



CD 

CD *0 

G c 




cd v CN 



0> 



3 ^ 60 60 , 

S S5 C C cr7 

£7n tD CD g 

H- CD ID CD 

^ 03 C C os 

pfi u C C 



110 



WISEMAN 



Vol. 12, No. 1 



T3 

c 



o 







Q 



c 

c 



CL 




cd 

O 



c 
£ 

QJ 

s 



c 

o 

X 
OS 

H 






>> 



03 

CD 

o 



cr 

0> 



3 



CV 



Q 



ex 

*3 

c 

CO 

C 

03 
> 

o 
o 

1-4 




CN 



C 

3 
E 

ON 




CN 



2 ^ 

C oo 



D 




I 

0> 





c 



cd 

35 



oo 





<3 

CJ 

CD 



U 



c 
3 




2 




• 




c 




*J 


• 


v 


\D 


CD 


00 


3 


in 


cd 


^— H 


v-« 




A3 


*-T 


O 


3 


2 


£ 


43 


v& 


* 9 


OS 


fc 


9 • 


cs 


^ 


m 


in 


m 


fH 





CD 



CD 

3 OO 

in 




oo 



CD 

3 



CD 



£ - 

O T3 

a; <; 



CJ 



3 

Q 



cu 

s 

CD 

a 



CD 

3 

03 

C 

c 



CD • 




<3 tJ 

e 

n3 



CD 

5J 



S>0 



O 



T3 

C 

CO 
CV 

n3 



CD 

8 



i 



o 



ftJ 



cj 






CD 

3 


CD 

03 



QJ 



CD 

85 





CJ 

ex 

en 






> 




T3 

03 




73 



sx * 



cs 



X 
* 



K * nJ ^ ,g 



CD 



J5 






^ 

^ 



y 

03 




A3 

ex 
£ 

a 

C 

QJ 




CX 



C~> 
CD 

CJ 

'So 






CD 



T3 

C 



§ E 



CD ^^ 
CD . v 

O ^ 





o 



« 



1 



CD 

X) 

o 

cy 

c 

£ 






c £ 

(T5 CD 



cjO 

c 

(T3 



CD 

o 
3 



«3 

3 

3 



o 



3 

CD 

c 





CO 

c 

CX 

O 



J- 

03 



fC 



A3 

CD 






o 
C 

c 

CD 

03 



g 

CD 



CD 
Q 

£ 

CD 



c 



o 

O 
CX 

C 

03 

.a 

CD 

c 



C 

CD 



03 






C 

o 



03 

e 

C 

o 



03 

u 



SJ 







CD ^ 



3 

03 rv 

U H 

<0 hJ 



TJ 



S 8 

^^ # ■ 

in m 

35 



CX 

CD 

CD 
3 




03 

T3 




o 



03 



o 



CX 

CD 

CD 



nJ 




cx.§ 
2 ^5 




o 
cn 



i- 3 






Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



111 



2 co 
O w 



co 

O ^ 



T5 

01 OS 

03 o 

o 1 

CD 



o 



n en 

5 A 



i 



c 




X 



(0 






03 



5 o 73 



s s 




S 2 « c 

u u o "^ 

5 C Vh !> 

U S 0) H 



OS 

y 3 



co 

CD 

6 



co 

a; 

«3 



.a o 
jo 



C 03 

OJ CD 



W3 « 

C a) 

^J 0J 



— nH 

S .2 <d § 

co 3 C o 

O &D;3 K 



CD 

cd qj 

■si 



co 




C 




C IN 

CD ON 



CD 

M CD 



cd 






as 

to 

5 



SK 



CO 

00 



a> 




CD 



i 



T3 

(D -a 
> 5 



CD 



lO 



1 = 

3 o 



X 

CD 



E = CO 

O X 03 

C ^ H 

CO r* 03 



CO 

co y 



T5 

» c 55 

P OS (Q 73 

~ C 



^ ^ #« 

^ -»-» CD 

CD cx 



g fc 



CO 



-c y w cd 

CD (3 

o -g ■ M 



^ 

u 



fa CD CD 03 

CO 3 <D X 
— twm * CO 



CO ;j w 



CO 

3 

O 

O 

CO 



in 



CO 

3 

CO 
CO 



03 



CD 

g 

C 

CO 

03 

H 




£2 J 

CO 

.2 £ 

03 



CO 

m 



V • 




CO 

B 

CD 

i 



03 Oh 

o 



00 
00 



I 



t-N. 



03 

C 



CD tJ< 

s ^ 

o 

& 

03 
U 

CO 



s 




ON 



CO 

00 
(N 



CO 

a. co 

U v 

03 m 

o g* 
ex « 






05 kJ 






03 



2 -J 

o S •* 

5 

• CJ u 

9«^ 



XI 



O 

u 

03 

O 



CO 
CO 

03 



3 
fi* -ri 






O 



x> 



o 

03 

O 



J 



ro 



03 CD 
U 



o 

3 

IN 
IN 

m 

I 

< 



U1 



-d 



C/5 

g 

CD 

e 




o 
SO 



« 



^ CO 

00 

00 



73 



I 






ON 

CO 

00 



03 

o ^ 



2 



• • • • 

CT^ O 

in m co 



CO 
CO 

03 



03 



CD 

£ 



CO 

03 



co 

h 

M 

CD 

S j 

o 2 .. 

y x in 



3 



m 




CO 

CD 

g 



o 

CO 

CO 

03 



2 (N 

co nJ 

H 

03 co 

3 

CD 

fit 



Pi 



T3 



O 

03 

o 



CO 



♦ • 



3 

c 

03 

£ 




00 






CO 

CD 



its 

£ 

3 



6 



» 



CO 



03 Q 
73 ^S 



C 

M N 

3 

U 



CO 

3 

CD 

03 

C 



J 




<D 

03 



03 



CO 

Oh 



CO 

3 

CD 

c 



^J 



s 




s 



^ 



CO 

fit 



CO 

3 

CD 

g 



2 00 



&"« H 



C 



^C 



Wi O C 

o TS 



CO 

CD 



CD 
03 



6p 



CD 



c7^ 



•8 

^ 03 




^ 



O 
2 



112 



WISEMAN 



Vol. 12, No. 1 



- 



C 
O 








c 




c 





a. 

D 



c 



C 

o 

X 



60 

E 

o 

sfc 

c 



LO 

J ° 



I 

c 

CD 



en 



CD 
CD 

03 



Q 



o 



00 

in 
oc 



CD 

O Q 

cj 



in 
in 



m 



in 



CD 
CD 




O 



in 



C 

03 

03 



<s in 



in 



CD 

2 

6 

3 



as 

£ 

o 

CD 




E 

3 



o 



CD 



CD 



ft « 



C 

03 

■ * 



OS 




QC 



03 





CO 
IN 



<d 




5 
<3 



c 

re 




.2 



Oh 

§•2 
^3 



03 .. 




m 
m 

oo 



CD 

5 

cp 



as 



C 





CD 

3 
u 

03 CD 

P *- 



B- 



c/> 




1/ 

OS 

T3 



00 

in 

IN 

CD a> 

£> OS 

33 






60 

C 



0) 

c 

OS 

£ 

ex 





CD 
CD 

03 C 

c *L 

03 ^ 



CM 



CJ 




CD 

3 

CD 

03 



c 



Q 




-Si 



1 C 



O 



60 

03 

£ 
o 




£ 

o 

ifi 

CD 

2 

£ 

3 



OS 

O 

CD 

3 

O 

o 

ex 



ij . 



CD 



1 

03 

o 

OS 
CD 

*<3 



on O 

CO 




CD 

3 

CXI 

£ 

3 



m 




^ 



T3 

to 



in 

00 

CN 

in 



CD 

CD 



CD 

s- 
03 



6- 



o 



CD 

- • 

ON 

in 



x> 



in 
tN 



3 ^ 
£* 



73 a; 



1 



< 



c 



Di 



T3 

C 

03 



w 00 

COO 



OS 
T3 



o 
U 



ll 

>» CD 

<-> e 
J^ 1 



u 

00 



. ^ 



O 

u u 






Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



113 



T3 

C 
3 




O 
PQ 



ID 
CO 



^O 



•J 



Q 



in 

3 



OS 



o 



ra 




o 

ON 

in 



co 



£ 

o 



E 

2 



c 
<5 



m 

CN 




2 8 
60 a. 




<N 



S. r- 



CO 








OS 

C 

05 

Q 



i 



6 



X) 



Q 

CO 

3 
co 



C 
2 



hJ 




in 
in 

CM 

m 

CM 
i 

< 



O 

co 

3 

co 



K 



CN 

c 

as 

s 

13 



Oh 




CN 
CD 



IS. 

ON 




s » 



2 rf 

2 £ 

Cl.cn 



00 

i 



CO 

o 



ON On On 

T^ ^ Tj< 

m m in 



i 



522 



On 

CO 

NO 

C <* 
• v 00 

ON IS. 

CO o 

NO 



3 



CO 

o 

ON 

in 



2 



o 

c 
o 



o 

ex 



£ £ 



3 



CO 

O 

ON 

in 



< 




CM 




CO ^o 

CN 



CO 

0> 



to 

E 



QJ 



-9 -S 



c 

s 



8 



CN 

O 

CO 

• * 

ON N 
t^ CN 

in m 

CN 

i 

< 



6 
g 



CO 

C 




O 

o: 

G 

3 
C 

£ 

CM 

ON 
CO 

vO 

@ 

a; 



CO 



CO 




C 

oj 

c 




IN 

CN 

ON 



CN 



E 

0> 



ON 
ON 

Q on 

s ^o 




cnj m 



CO 



CJn 5 
NO <N 



CO 



J sC 



CN 

ro 
fN 



£ 

3 



c 

35 

CO 






OS 



£ 




CO 



3 s 

'^ CO 

CO OC 

CM 



c? 




ON 

£ 

3 

CO 



CO 

3 
01 

00 3 

. OS 




0> 

E 

3 



I* 






& 



fN 




cx 

CO 

CO 



CO 



Q 

c 



"!3 



T3 



5 
co 



N 

o> 
Xi 

£ 



q 

c 



Ts 




00 

in 



CO 

3 

0» 
05 

C 

c 



J 



CO 

5 



fc 

s 



co CO 
<u CN 

CO °° 




5 




U 



a 

CO 




u 

c 



O) 

C 



>^ 2 

en ^ 







£ 



CO 



5 

O 




NO 

vO 
IN 



§.£^ 



CO 

3 



3 
^3 



CO 

3 

OJ 

C 
C 



ooL] 



co 




3 



^o 



CO 



8- 



jr on 

Q CN 

co ^ 

^ 5c 

05 



o> 

C 




a; 

05 

C 

o> 



u 



0> 

no 
C 



C 

CO C 

0» ^ 

£ aJ 

° 6 

CO MM 

CO 



114 



WISEMAN 



Vol. 12, No. 1 



73 

C 



c 

o 









c 





'5 

CD 

Q 



c 

cd 

S 



c 

c 

X 

A3 



On 



^ 



On 

IN 

CO 



CD 

3 






T3 



^ 1 



E 

CN 

do" 



ON 

CO 

CD 
03 

c 





IN 



O 3 




CN 



nO 



00 

ON 

SO 

in 

ON 
ON 



NO 

no 

ON 



CN 

NO 

in 



CO 

00 

ON 

CO 

NO 




cm 



00 
IN 



On" 

00 



CD 

1 

CD 

E 

"9 H 



oft 



cd 



ON 

GO 



ON 

CO 

ND 

CD 

3 



in 

oo 



in 




c 

CD 



C 









ON 

CO 

vO 



in 

00 

ON 

CO 

00 

ON 

CM 

00 



cd 

03 

c 
3 



t> ft 



CO 

in 



m 

^o 

m 

z 

oo 

CM 

m 

co v 
m 
in 



tN 



so 

NO 

in 



CM 




c\i m 



IN 

ON 

GO 

CM 

00 



On 
IN 



m 

oo 

<D 

03 

C 
cN 



:z 

in 

CN 

m 

5 



00 

SO 

GO 

CD 
03 

c 



CM 



m 

CM 
CM 

ON 



CM 



(N 



3 CN 



O ON 

00 CN 

m > 

CO «tf 

00 

On" 
CO 

NO 

oo v 

00 



CN 
CN 

IN 

00 
IN 



i 
CD 

E 

o 

<J 

NO 



CO 




CM 

ON 
ON 



I 

<D 

£ 

o 

Oh 

03 



cb 






I 

O 




IN 



03 
U 
03 



CO 



6 ^5 



in 

CO 

m 

z 

CO 



ON 

CO 

NO 



J2 




CD 
CD 

H 
03 





CD 

03 



00 



00 



CO ^J 

cm 





rH 




IN 

CN O 

m 
i 



00 ^ 



5 

CD 



CO 
03 





ON 



03 



00 








CO 

in 



NO 
NO 
ON 



in 

oo 



IN 

ON 

CO 



CD 

o: 

oq 3 




CD On 





CD 
03 

' C 



e 



tN 9 






Q 



^QO 

03 

CJ 00 

N-J 00 



W 

NO 

\o 
m 

Z 

00 
CN 

in 



CO 
00 

ON 

CO 

NO 

CO 

CN 





o 

tTON 
03 CO 

KJ NO 



in 

ON 

CO 

NO 



o 



jO 



I 



ON 





Oh 

ON 



03 

o 

6 

CD 



5- 

CD 






m 



E 

CN 

ON 

CO 

NO 

CD 
CD 

03 



i 




00 



O 
2 



m 

CN 



CD 

6 

o 

CD 

03 






03 

U 
03 



w 

NO 

NO 

in 

Z 

oo 

(N 

m 



CD" ""J 

S oo 

03 Es 



O 
2 





03 

O 

6 

CD 



• ex 

in 



in 

CM 

in 

oo 

00 
On O 



CN 

oo" 

NO 



CO 




ON 




CD 

H 
03 




O 



i 

m 

on 



cr> 

CD 

i- 

03 



a q 




x> 



Q 

CO 



CN 

m 

O Tf 



CD 




00 



w 

NO 

vO 

in 
in 

CN 

m 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



115 



I 

03 



0> 

6 

o 

cd 



03 



& 



CO 

en 

CM 
CM 

^C 



^ 



LO 

O CN 
CM 



ON 

CO 






CM 



5 



cn 



in o 

00 On ^. 
LO r-» <N 



CD 



CO 



00 
On 



£ 

o 

CD 

03 



CD CD 

J- *- 

03 OJ 



Oh 



co 




in 

oc 

CD 



ON 

co 

NO 



CD 



03 



CO 



i 

o 

CD 

'— 

03 



Q 

ON 
ON 



in 

cn 



CD 

3 

CD 

s- 
03 



03 
C 

3 

C 

03 



oCT3 

CO 



00 



NO 



"S 5 r^ 

o 

CD 

H 

OS 



o 

u 

03 





£ .^« 



o 

CD 

03 



CD 

03 



- • 



IN 
IN 

03 <NI 






a> cm 



£ 



i 




O 
C 

Q 

C 



II 



43 

•c 

I 

ii 

i 

o 
ex 

II 



re 

In 

II 

Q 



116 



Vol. 12, No. 1 



/. Ethnobiol. 12(1): 117-139 



Summer 1992 



FOOD TABOOS AT BUZIOS ISLAND (BRAZIL): 
THEIR SIGNIFICANCE AND RELATION TO FOLK MEDICINE 



ALPINA BEGOSSI 

Nucleo de Estudos e Pesquisas Ambientais 
Universidade Estadual de Campinas, CP 1170 

Campinas, 13081, SP, Brazil 



ABSTRACT.— Buzios Island is a fishing community of about 44 families located 
off the coast of Sao Paulo State, Brazil. This study focuses on the food taboos 
of the islanders from Buzios, specifically on prohibitions in the consumption of 
certain species of fish and lizard. Interviews and observations on food habits were 
made from November 1986 to December 1987. Some fish are avoided only when 
persons are ill and others are recommended for these cases. Among the generally 
avoided fish, moray and ray receive special attention. The reasons for these 
avoidances were investigated and it was observed that carnivorous fish, more 
than other fish, are avoided during illnesses. The toxicity of fishes might also 
explain some avoidances. Finally, medicinal uses of some species, such as rays 
and lizards, seem to explain some important dietary taboos: medicinal animals 
may be saved as a source of drugs. 

RESUMO.— A Ilha dos Buzios, situada no litoral do Estado de Sao Paulo, e 



44 



alimentares 



de Buzios, especialmente aqueles referentes ao consumo de certos peixes e do 
lagarto. Entrevistas e observances sobre habitos alimentares foram realizadas de 
novembro de 1986 a dezembro de 1987. Alguns peixes sao evitados, enquanto 
outros sao recomendados em caso de doencas. Dentre os peixes evitados no con- 
sumo alimentar merecem atencao a moreia e a raia. As razoes dessas proibicoes 



carnivoros 



almente, o uso medicinal de alguns 
animais, como da raia e do lagarto, parece explicar alguns tabus alimentares 



importantes. Animais usa< 
como fonte de remedios. 



RESUME. -L'ile de Buzios, situee au litoral de Sao Paulo, est peuplee par une 
communite* de pecheurs, approximativement 44 families. L'objectif de cette etude 
est de comprendre les prohibitions ("taboos") alimentaires des habitants de 
Buzios, en particulier celles qui se rapportent a la consomation de certains poissons 
et du lezard. Des entrevues et des observations sur les habitudes alimentaires 
ont ete realisees de novembre 1986 a decembre 1987. Certains poissons sont evites 
et d'autres sont indiques en cas de maladie. Parmi les poissons evite's, dans la 



consomation 



ces prohibitions alimentaires sont analises. Les poissons carnivores sont, en 
ge'ne'ral, evite's par les malades. Les poissons toxiques peuvent expliquer d'autres 
prohibitions. Finalement, l'usage medicinal de certains animaux, comme la raie 
et le le'zard, semble expliquer certaines prohibitions alimentaires. Les animaux 
utilises dans la medicine menagere pourraient etre epargne's, entant que source 



de medicine. 



118 



BEGOSSI Vol. 12, No. 1 



INTRODUCTION 



Food taboos and preferences have been an area of debate in ecological-cultural 
studies between materialists/utilitarians and symbolists/ structuralists. Vayda (1987) 
reviewed part of this debate through the analysis of Harris's (1985) cultural 



materialist 



ecological 



to the main animal avoidances reported and observed at Buzios Island, and to 
verify the relationships of the patterns of avoidances to protection of medicinal 

animals. 

There are different questions concerning food taboos and the schools of 
thought mentioned above reflect these approaches. Symbolists have studied the 
relationship of taboos to religion and rituals and have focused on their emic aspects. 
For example, Barthes (1961) analyzed the psycho-sociological aspects of certain 
contemporary American and French feeding habits. Douglas (1969), in the classical 
study "Abominations of Leviticus," concluded that dietary laws were liturgical 
signs of purity. Sahlins (1976), stressing the importance of "cultural reason" to 
explain human food habits, states that edibility is inversely related to ' 'humanity 
For example, American avoidances of horse and dog are explained because they 
participate as subjects (and not objects) in American society. Dogs and horses 
are named and people commonly converse with them. On the other hand, this 
behavior does not occur with edible animals, such as pig and cattle. Another emic 
approach was taken by Basso (1972) in her study of food classification and linguistic 
categories among the Kalapalo Indians of the Upper Xingu (Brazil). 

The questions which interest materialists are essentially related to environ- 
mental aspects of the issue and represent an etic approach to dietary practices. 



// 



comprehension of indigenous ideologies may 
material circumstances. For example, Harris ( 



1987b) attempt 



human 



taboos and preferences. Harris's analysis of the taboo of the sacred cow of India 
indicated that cows were too important to be eliminated. They were animals that 



farmers who 



1977 



Other studies have included both emic and etic analyses, such as Ferro-Luzzi 
r 5) study on the positive and negative attitudes towards food among India 



may 



McDonald (1977) and by Reichel-Dolmatoff (1976). The 



important point, as mentioned by Basso (1978), is that these different perspec- 
tives on food taboos and preferences are not incompatible but complementary. 
The approach taken in this study is both etic and emic. The strong etic basis 
is reflected in reference to aspects of the environment, such as fish diet, fish 
toxicity, and medicinal use, that might explain some food taboos. Emic aspects 
include the perceptions islanders have about fish, such as their behavior or smell. 
The aim of this study is the search for possible environmental reasons that could 
be behind the food taboos observed at Buzios Island. 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 119 



There 



Amazonian 



in particular by Ross (1978). He stressed that Amazon food prohibitions can be 
explained by the costs and benefits of resource procurement strategies. Most data 
on Amazon food taboos concern game animals (Ayres and Ayres 1979; Chagnon 
and Hames 1980; Kiltie 1980; Moran 1977; Smith 1976). Food taboos related to 
game and fish species are reported by Basso (1972, 1973), Moran (1974), and Pereira 
(1974). 



There 



Amazon. Gouldine (1981) observed 



fishes (gymnotoids) are the third most abundant group in the Amazon, they have 
no importance in the fisheries because most are small and stay hidden, and there 
are cultural taboos against eating them. Smith (1981) pointed out that certain fish 
species are avoided by Camayura and Tapirape Indians, as well as other Amazon 
inhabitants. Among Indians from the Upper Xingu, fish avoidances related to 
the physical state of a person and to the diet of fish were observed (Basso 1972). 
Recently, Begossi and Braga (1992) analyzed the relationship of food taboos and 
fish diet on the Tocantins River. 



THE COMMUNITY 



Buzios Island is a fishing community of about 220 individuals (44 families) 
located on the coast of Sao Paulo State (Fig. 1). The population of Buzios is 
distributed among eight small harbors, each of which has a dock and canoe 
shelters. The most copulated harbor is Porto do Meio which included 23 families 



time 



from 



They 



Tupinambas 



main 



manioc 



(Mussolini 1980). In Brazil, slavery 



m 



nineteenth century and African features are found in the 
igical beliefs of the caigaras, such as in the festival of Sao 
Benedito, a black saint (Correa 1981). In the beginning of this century, Japanese 
migrants introduced to the coast of Sao Paulo a type of fishing trap, the cerco 
flutuante, which is made of floating chambers of net (Japanese: kaku-ami, Von 
Brandt 1984) (Mussolini 1980). This fishing technique is also used at Buzios Island. 
In the past agriculture was important at Buzios (Willems 1952). During the 
field work reported here (14 months in 1986-1987), islanders spent most of their 
time in fishine activities. Manioc is cultivated by some families, usually for home 



(Begossi 1989a) . 



tein consumption comes especially from fish. The typical meal 
squid with manioc flour, rice, beans, and sporadically, spaghetti 



120 



BEGOSSI 



Vol. 12, No. 1 




FIG. 1.— Map of Brazil showing 
Paulo State. 



METHODS 



The field work at Buzios Island included observations and interviews 
formed during monthly visits of about a week each from November 191 
December 1987. A total of 88 days was spent at Buzios Island as the guesl 
fishing family. I sampled the diet of 10 families from Porto do Meio duri 
12 months period (5 davs/month): a detailed studv on fish nrpfprpnrps at Bi 



in 



interviews were carried out as part 



hshing at Buzios Island (Begossi 1989a). Interviews were made at six of the eight 
small harbors of the island: Porto do Meio (23 houses), Guanxuma (7 houses), 
Pitangueira (4 houses), Costeira (1 house), Porto do Cais (2 houses), and Mae 
Joana (2 houses). The two other harbors were not included in the study due to 
difficulty of access. Interviews included questionnaires and were conducted when 



women 



interview was made with 73 islanders (about 88% of the adult residents). This 
interview included the questions "Que peixes voce nao come?, Porque? (What 
kind of fish don't you eat? Why?). Another interview, with 57 people, included 
questions such as "Que peixes voce nao come quando esta doente?" (What kind 



Summer 1992 JOURNAL OF ETHNOBIOLOGY U l 



ish don't you eat when you are sick?) and "Que peixes voce come quando 
i doente?" (What kind of fish do you eat when you are sick?). Questions 
cerning medicinal animals were asked in another interview, in which 32 
riders described animals used to cure diseases or heal wounds. Informal 
versations and direct observations on food taboos and folk medicine were made 
'orto do Meio harbor. These included observations on foods avoided by 
nders and on the preparation of home-made medicines. 
Analyses of the lipid contents of fish species were based on samples of 



Guimaraes of the Laboratory 
Campinas, Sao Paulo. Details 



made by E.S. Contreras and J 



Fish collected at Buzios Island were identified using the keys of Figueiredo 
7), Figueiredo and Menezes (1978, 1980), and Menezes and Figueiredo (1980, 
>). Identifications of molluscs were based on Rios (1985). Specimens collected 
e deposited at the Museu de Zoologia, Universidade de Sao Paulo (MZUSP). 



FISH TABOOS 

The fish most strongly avoided as food by islanders from Buzios are camburu 
(moray, Gymnothorax spp.), raia (species of ray), bonito (bullet mackerel and 
little tunny, Auxis sp. and Euthynnus alletteratus) , and tinhuna (sargeant major, 
Abudefduf saxatilis) (Table 1). Begossi (1989a) used regression analysis to investigate 
the variables (calories, fish boniness, price, and availability, among others) 
involved in the choice of animal protein, consisting mostly of fish, by islanders 
from Buzios. Fish avoidances or taboos could not be explained by any of these 
variables. 

The reasons interviewees offered for avoiding fish varied among fish species 
(Table 2). For example, the avoidance of camburu (moray) is explained by its 
"snake-shape," bad smell, aggressive behavior, ugly appearance, and con- 
spicuous teeth (Fig. 2). Ray avoidance had similar explanations, except that rays 
are not "snake-shaped" nor do they have teeth. In fact, aggressive behavior may 
help explain the avoidance of certain prey types. A fisherman at Buzios who 
caught three camburu (moray) had to leave two at sea due to the difficulty he had 
removing the hooks from these aggressive prey. Lenko (1965) reported that 
inhabitants from Buzios avoid eating camburu because they believe it climbs out 
the sea to fight the poisonous jararaca (Bothrops sp.), a very common snake on 
the island. Bonito (bullet mackerel and little tunny) (Fig. 2) is avoided due to 
presence of "blood" (i.e., a high concentration of hemoglobin) (Moyle and Cech 

1982). Tinhuna is said to have a strong or bad smell. 

Some fish avoided by islanders, such as bonito, ray, and shark (Table 2), are 

termed carregado. Carregado includes a set of supposed attributes of an animal, 
such as teeth, blood, aggressive behavior, "strong flesh," fattiness (graxa) , and 
factors that could cause inflammation if eaten by someone who is wounded or 
unhealthy. A frequent answer during interviews was "se comer peixe carregado 
a ferida inflama" (if you eat a carregado fish, you will get inflammation of wounds). 
Women are not supposed to eat these fish during menstruation or after child- 
birth. 



122 



BEGOSSI 



Vol. 12, No. 1 



TABLE 1.— Fish avoided by at least 3% of interviewees (total number: 73) from 
Buzios Island. 

Names Percentage 

of interviewees 

Local English Scientific practicing avoidance 

camburu moray Gymnothorax spp 25 

raia ray Raja cyclophora, Myliobates sp., 

Dasyatis sp. (?) 21 

bonito mackerel, tunny Auxis sp., Euthynnus alleteratus .. 18 

tinhuna sargeant Abudefduf saxatilis 16 

cagao shark Rhizoprionondon lalandei, 

among other spp 15 

corvina sand drum Umbrina coroides H 

garoupa grouper Epinephelus spp 10 

piragica yellow chub Kyphosus incisor 8 

bagre catfish Notarius grandicassis 7 

budiao hogfish, parrotfish Bodianus spp., Scarus vetula, 

wrasse Sparisoma spp., Halichoeres spp. ... 7 

enchova bluefish Pomatomus saltator 7 

salema porkfish Anisotremus virginicus 7 

bicuda guachanche Sphyraena guachancho 6 

goete weakfish Cynoscion spp 6 

betara kingfish Menticirrhus americanus 4 

espada cutlass fish Trichiurus lepturus 4 

frade angelfish Pomacanthus paru 4 

tainha mullet Mugil plat anus 4 

baiacu puff erf ish Sphoeroides spengleri 3 

corcoroca tomtate Haemulon aurolineatum 3 

matnangaba scorpionfish Pontinus rathbuni 3 

olhete yellowtail Seriola lalandi 3 

pargo spadef ish Chaetodipterus faber 3 

sargo black margate Anisotremus surinamensis 3 

xalerete bluerunner Caranx crysos 3 

small and rocky fishes 3 



Observations: 10% of interviewees mentioned they eat any kind of fish; 4% r 
they usually avoid eating turtle (Chelonia mydas); another 4% avoid eating dolphin 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



123 



Moran (1974) observed that Amazon caboclos (people of mixed European and 
Indian background) have some food restrictions during illness, pregnancy, lacta- 
tion, and menstruation. The foods prohibited are called hot (quente or reimoso). 
Smith (1976, 1981) studied food habits of transamazon settlers and of caboclos from 
Itacoatiara in Amazonas State. In these cases, the fish avoided, referred to as 




FIG. 2.-Some animals avoided at Buzios Island: (a) a fisherman 



catnbu 



moray, Gymnothorax funebris); (b) some Scombridae 
ullet mackerel, Auxis sp.) (top) and bonito-pintado ( 



s) (bottom) 



124 



BEGOSSI 



Vol. 12, No. 1 



TABLE 2.— Explanations given by Buzios islanders about fish avoidances. 



Explanation 



Fish 



In case of disease 



Avoided (n = 65) 



Avoided (n = 32) 



Eaten (n = 8) 



Never ate it baiacu, boto, camburu, 

raia, tainha, turtle. 

Snake-shape camburu 

camburu, frade, 
garoupa, piragica, 
raia, tinhuna, cagao 

Dangerous, camburu, raia, caqao 



Bad smell 



aggressive 

Carregado 1 

Blood 



bonito, raia, caqao 
bonito 



All cited in Table 3 
bonito 



Illness, body bonito, bicuda, cavala, All cited in Table 3 



wounds, 
toothaches, 
after child- 
birth* 



enchova, espada, goete 
raia, tainha, sororoca, 
cagao 



Ugly, nasty camburu, raia 



Fatty 
Scaleless 



garoupa, goete 

caqao 



olhete 



Many bones budiao, corcoroca, 

jaguaricd, small 
rocky fish, sardinha, 
tinhuna 

It eats mud corvina, timbali 

or dirty 
things 

Hard flesh 

Soft flesh 

Toxic 



Sting 
presence 

Difficult to 
capture 

Bad for 
women: it is 
believed 



woman 



corvina 
namorado 
baiacu 
mamangaba 



raia 



raia 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



125 



Explanation 



Fish 



In case of disease 



Avoided (n = 65) 



Avoided (n = 32) 



Eaten (n = 8) 



explodes, 
visceral 
organs are 
exposed 

Conspicuous 
teeth 



Weak flesh3 
With scales 



Docile, not 
aggressive 



Little blood 



bicuda, camburu 
enchova, espada, 
marimba, piragica, 
paru, cagdo 



Tooth absence 



par go, piragica 

garoupa, marimba, 
piragica, panaguaiu 

garoupa, jaguariga, 
marimba, olhete, 

piragica, salema 
sargo. 

badejo, garoupa, 
marimba, olhete 
cagao 

ca$ao 



^Carregado means with conspicuous teeth, fatty (graxa), with blood, aggressive, with 
strong flesh, or that it exacerbates any illness. 

2 Body wounds are the most cited health problem associated with the consumption of 
some fishes. 

^Opposed to strong flesh or carregado. 

Scientific names of fishes absent from other tables (mentioned by only one interviewee): 
cavala (Scomberomorus cavalla), namorado (Paranthiasfurcifer), sardinha (Sardinella brasiliensis), 
timbali (Fistularia petimba). 

Observations: Another interview, only about ray avoidance, showed that 63% of inter- 
viewees (n = 30) do not eat ray, for the reasons listed above. In the interview reported here, 
explanations such as "I do not like it" are not included; scientific names are found in 
Tables 1, 3, and 4. 



reimoso, are those that cannot be eaten by anyone suffering from a wound, 
measles, tumors, and skin rash; these fish are believed to exacerbate health 
problems and to have a "strong meat." Pereira (1974) observed that scaleless 
fish (peixes de couro) are usually avoided in the Amazon River and that some 
illnesses are believed to be caused by these fish. Begossi and Braga (1992) observed 
similar fish avoidances bv fishermen living along the Tocantins River. These 



126 



BEGOSSI Vol. 12, No. 1 



fishermen also called the avoided fish reimoso or carregado. The similarities of names 
and meanings of avoided fish in regions so distinctive environmentally as the 
Amazon and Buzios Island are worthy of investigation. 

Similar food taboos have been reported in other regions of the world. Wilson 
(198(ft reported avoidances of ray, bonito, and mackerel, among others, by Malay 



women 



same genus asbicuda), and Scomberomorus (same g 
pregnant and puerperal women from Talminad 



mi 



similar fish avoidances from such diverse communities 



DIET RESTRICTIONS IN CASE OF DISEASE 



Among the fish avoided, interviewees emphasized that some 



mainly by people suffering from disease or wounds 



interviews 



The most important 



mackerel 



) and enchova (bluefish, Pomatomus saltator) 
pted or even recommended in case of illness. The 



most important of these are marimba (spottail pinfish, Diplodus argent 
garouva (grouper, Evinephelus spp.) (Table 4). Caqao (small shark) was m 



recommended 



ar 



among 



Some diseases, such as bronchial asthma and psoriasis, as well as heart 
attacks, are not as frequent in populations with a fish-based diet, such as the 
Eskimo (Lands 1986). The consumption of different species of fish might also have 
effects on the health of consumers. 

Table 2 shows that many fish avoidances are explained by Buzios islanders 
by the fact that the fish are carregado. According to Smith (1981), reimoso fish are 
considered to be oily; many catfishes, for example, have abundant fat reserves. 
High fat concentration was cited by interviewees from Buzios as the reason to 
avoid a few types of fish (Table 2). Data on lipid contents of fish species men- 
tioned by more than 10% of interviewees are shown in Table 5. There is no signifi- 
cant difference between the mean fat content of fish which are avoided and those 
which are eaten (ANOVA I, F, p > 0.77). Thus, fat content cannot explain why 
some fish are avoided and others are preferred as food in case of illness. Begossi 



)ided by fishermen living along the 
consumed and preferred bv them 



high in calories. 



dur 



The fish comm 



are considered docile (Table 2). As shown in Table 6, fish which are avoided by 
unhealthy persons are usually predators of other fish species. On the other hand, 
most fish accepted during sickness feed on invertebrates or plankton. Thus, the 
diet of the fish seems to determine which species islanders avoid and accept as 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



127 



TABLE 3. 



Buzios for fear of exacerbating disease. 



number: 57) 



Names 



Local 



English 



bonito 



enchova 
xalerete 
espada 
cagao 



mackerel 
tunny 

bluefish 

bluerunner 

cutlass fish 

shark 



bicuda 
sororoca 
goete 
raia 



guachanche 
mackerel 
weakfish 
ray 



tainha 



xareu 



jaguartga 
camburu 
olhete 



mullet 
jack 

squirrel fish 
moray 
yellowtail 



Percentage 

of interviewees 

Scientific practicing avoidance 

Auxis sp. 

Euthynnus alletteratus 74 

Pomatomus saltator 68 

Caranx crysos 35 

Trichiurus lepturus 30 

Rhizoprionondon lalandei, 

among other spp 25 

Sphyraena guachancho 14 

Scomberomorus brasiliensis 11 

Cynoscion spp 9 

Raja cyclophora, Myliobates sp. 
Dasyatis sp. (?) 7 

Mugil platanus 7 

Caranx latus 7 

Holocentrus ascensionis 5 

Gymnothorax spp 4 

Seriola lalandi 4 



Observation: Nine percent of interviewees 

ill. 



food when they are ill. Invertebrate or plankton feeders are said to have "weak 
flesh/ 7 i.e., a kind of meat that does not exacerbate health problems. Smith (1981) 
observed that fish considered reimoso are those which eat all kinds of creatures. 
Most fish avoided by people from the Tocantins River region are carnivorous 
whereas consumed and preferred fish are usually herbivorous or detritivorous 
(Begossi and Braga 1992). Some fish can be especially unhealthy if they occa- 
sionally eat other venomous fish. The probability of acquiring (and accumulating) 
toxins increases in higher trophic levels. These toxins may be pollutants, such 
as heavy metals, or natural substances, such as those obtained by eating toxic 
plankton, invertebrates, or fish. 



TOXIC FISHES 



Most families at Buzios consider baiacu (puff erf ish, Sphoeroides spengleri) 
venomous (Lenko 1965). Pufferfish poisoning has been reported since the s 
teenth century (e.g., Piso 1658). Actually, tetrodoxin, a neurotoxin, is pi 



128 



BEGOSSI 



Vol. 12, No. 1 



TABLE 4.-Fish favored by at least 3% of interviewees (total number: 57) from 
Buzios during illness. 



Names 



Local 



marimba 
garoupa 
piragica 
caqao 



sargo 

panaguaiu 

tinhuna 

salema 

cor cor oca 

jaguariqd 

olhete 

badejo 

budiao 



olho de box 

pargo 

xalerete 



English 



spottail 
grouper 
yellow chub 

shark 



black margate 
halfbeak 
sargeant 
porkfish 
tomtate 
squirrel fish 
yellowtail 
black grouper 

hogfish, 

parrotfish 

wrasse 

amber jack 

porgy 

bluerunner 



Percentage 
of interviewees 

Scientific 

Diplodus argenteus 4 ? 

Epinephelus spp 44 

Kyphosus incisor 37 

Rhizoprionondon lalandei, 

among others 16 

Anisotremus surinamensis 16 

Hemiramphus balao 14 

Abudefduf saxatilis I 2 

Anisotremus virginicus H 

Haemulon aurolineatum 7 

Holocentrus ascensionis 7 

Seriola lalandi 5 

Mycteroperca bonaci 4 

Bodianus spp., 

Scarus vetula, Sparisoma spp. 

Halichoeres spp • 4 

Seriola dumerili 4 

Calamus penna, Pagrus pagrus 4 

Caranx crysos 4 



Observations: A total of 14% of interviewees don't consume any 
illness. Another 5% mentioned chicken as a food for ill persons. 



in liver, ovaries, and skin glands of pufferfish (Gopalakrishnakone 1988; Watabe 
et al. 1987). Species oibonito (Fig. 2), rays, and sharks are also avoided at Buzios 
(Table 1, Table 3). According to islanders, the meat of bonito deteriorates very 
rapidly. Mackerel-like fishes, tunas, skipjacks, and bonitos, if not adequately 
preserved, may be occasionally poisonous— a toxic substance is formed within 
the body musculature (Russell 1965)— and shark muscles have been mentioned 
as toxic, as has the liver of other elasmobranchs (Hashimoto 1979). Fish taboos 
of this kind have also been reported in India and Malaysia (Ferro-Luzzi 1980a, 
1980b; Wilson 1980) . 

More than 500 fish species have been reported to cause intoxication in humans; 
their distribution favors the tropical seas (Habermehl 1981). These fish tend to 
occur more frequently around islands than along continental shores (Russell 1965). 
Much of the available information on the distribution of toxic fish is based on 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



129 



TABLE 5.— Lipid content of some fish species from Buzios Island. 



Fish 



AVOIDED 

camburu 
bonito 



raia 



corvina 

tinhuna 

cagdo 

MEAN 



Gymnothorax spp. 
Euthynnus alletteratus 
Dasyatis sp. (?) 
Umbrina coroides 
Abudefduf saxatilis 
Sphyrna sp. 



Percentage 
Lipids 



9 

3 

1 
4 

3 

1 

4 



Total 
Solids (gr) 



64 

58 
63 
56 
72 
54 



AVOIDED DURING ILLNESS 



cagao 

bonito 

enchova 

xalerete 

espada 

sororoca 

bicuda 

MEAN 



Sphyrna sp. 
E. alletteratus 
Pomatomus saltator 
Caranx crysos 
Trichiurus lepturus 
Scomberomorus brasiliensis 
Sphyraena guachancho 



1 

3 

3 

13 



54 
58 
57 
61 



5 



ACCEPTED DURING ILLNESS 



cagao 

piragica 

marimba 

garoupa 

salema 

tinhuna 

panaguaiu 

sargo 

MEAN 



Sphyrna sp. 
Kyphosus incisor 
Diplodus argenteus 
Epinephelus sp. 
Anisotremus virginicus 
A. saxatilis 
Hemiramphus balao 
Anisotremus surinamensis 



1 
6 
5 
6 
7 
3 
2 
3 

4 



54 
54 

61 
58 
59 
72 
48 
59 



Sample base: 100 gr of salted and dried tissue (Begossi 1989a). Only fish mentioned by 



more than 10% of interviewees are included. 



130 



BEGOSSI 



Vol. 12, No. 1 



where 



ewis 



some toxic species is shown in Table 7. Many 



occur 



TABLE 6 —Feeding habits of fish avoided and eaten during illness by islanders. 



Fish 



AVOIDED 

bonito 

cagao 

enchova 

xalerete 

espada 

sororoca 

bicuda 



Family 



Feeding habit 1 



Scombridae 



fish and squid 



Carcharhinidae 2 carnivorous 



Pomatomidae 

Carangidae 

Trichiuridae 

Scombridae 

Sphyraenidae 



fish 

small fish: herrings and anchovies 

fish 

fish and squid 

fish and Crustacea 



EATEN 

cagao 

piragica 

marimba 

garoupa 

salema 

tinhuna 

panaguaiu 

sat go 



Carcharhinidae 2 carnivorous 



Kyphosidae 

Sparidae 

Serranidae 



invertebrates, vegetal matter 
algae, Crustacea, molluscs 
fish 



Pomadasyidae annelids, Crustacea, molluscs, ophiuroids 

Pomacentridae plankton, invertebrates 

Exocoetidae zooplankton 

Pomadasyidae Crustacea, equinoderms, small fish 



iData are from Figueiredo and Menezes (1980), Menezes and Figueiredo (1980) and Moyle 
and Cech (1982). 

2 This is the most common family; other families are Sphyrnidae and Odontaspididae. 



in Tables 3 and 4. 



interviewees 



Ciguatera has not been reported for southeast Brazil, nor are there known cases 
ish intoxication in this region. However, information is lacking on the food 
its of fishine populations and on thpir Pvnpripnrp wifh fnvinc Mqo fish men- 



interviewees 



man in other regions. As pointed 
i a particular locale may be cigua 
ly. Ciguatera poisoning is usually 



but not 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



131 






TABLE 7.— Some poisonous fishes. Data compiled from Rosenberg (1987) and 
Russell (1965). 



Fish name 



Local 



English 



baiacu 



barracuda 
bicuda 

budiao 



puffer 
bandtail puffer 

barracuda 
guachanche 

hogfish 
parrotfish 



wrasse 



cagao, tubarao dog fish 

spiky jack 



camburu 



hammerhead 
moray eel 



caranha 



snapper 



garoupa 



grouper 



jaguartga 



squirrel fish 



olho de boi 


amber jack 


peixe-porco 


file fish 


raia 


lesser electric ray 




spotted stingray 




stingray 




eagle ray 


xalerete 


bluerunner 


xareu 


horse-eye jack 




cardinal fish 




jack crevalle 



Scientific 



Occurrence 1 



Sphoeroides (14 spp.) G 

Sphoeroides spengleri S 

Sphyraena (7 spp.) G 

Sphyraena guachancho* S 

Bodianus rufus* S 

Scams (7 spp.) G 

Scarus vetula* S 

Bodianus bilunulatus G 

Squalus acanthias B 

Squalus fernandinus B 

Sphyrna zygaena* G 

Gytnnothorax (9 spp.) G 

Gymnothorax funebris* S 

Gytnnothorax moringa* S 

Lutjanus (28 spp.) G 

Lutjanus cyanopterus* S 

Epinephelus (8 spp.) G 

Epinephelus morio* S 

Holocentrus ascensionis S 

Seriola aureovitta G 

Seriola dumerili S 

Aluterus monoceros* S 

Alutera scripta G 

Narcine brasiliensis B 

Aetobatus narinari B 

Dasyatis (22 spp.) G 

Myliobates (6 spp.) G 

Caranx crysos* S 

Caranx latus S 

Apogon sp G 

Caranx hippos G 

Mugil cephalus* G 

Mycteroperca tigris* G 

Upeneus arge* G 



1 S: species collected at Bu'zios Island; G: genera found at Buzios; B: genera or species 

M m a -m i_ _ ^ ■ jt ^^X I^^^^^VV 



(b 



iguatera 



132 



BEGOSSI Vo1 - 12 ' No - 1 



Pacific Islands, on the coasts of the United States, and in the Caribbean 
(Habermehl 1981; Russell 1965). It is caused by the ingestion of fish with ciguatoxin 
(Hashimoto 1979). This neuromuscular toxin is acquired by fish through the 



bottom- dwelling dinoflaeellate Gambierdiscus toxicus (Lewis 



Miller 



from 



families 



Carangidae, Chaetondontidae, Labridae, Lethrinidae, Lutjanidae, Muraenidae 



implicated in ciguatera 



poisoning (Habermehl 1981) . The most common 



omnivores, such as Svhvraena barracuda 



ewis 1984; Quod 



The avoidance of top carnivores by islanders from Buzios (Table 6) could be the 
result of a cautious diet; when a person is unhealthy avoidances (taboos) are 
adhered to more strictly. According to Russell (1965), in cases of ciguatoxin, while 
both herbivorous and carnivorous fishes can cause ciguatera, the latter are more 
toxic and, in some areas, they are the only sufficiently toxic fish to cause poison- 
ing in man. Kodama and Hokama (1989) pointed out that, in cases of ciguatera, 
carnivorous fishes may concentrate toxins and even modify them biochemically 
leading to the formation of other variants of toxins. Moreover, other kinds of 
toxins which may result in fish avoidances may be still undetected. 



OTHER FOOD TABOOS AND FOLK MEDICINAL ANIMALS 



Among the available animal protein at Buzios Island, beef (usually dry meat: 
jerky), followed by fish and chicken are the items preferred by fishing families 
(Fig. 3). However, fish are the items consumed most frequently (Begossi 1989a). 
Dry meat is expensive and cannot be a common item of diet except for the most 
prosperous households in Porto do Meio Harbor. Among the animals avoided 



most important are lizard, octop 



4) 



food bv 96% of interviewees 



animal protein avoidances. The 



consumption 



pit 



animal. This behavior expresses how nasty and 



point 



main source of protein and fat for some human populati 

part of the Great Sandy De: 



Australia (Cane 1987) . Other populations, such as some from 
avoid eating lizards (Rea 1981). 

Lizard is a high caloric meat (293 kcal. in 100 gr salted and dried tissue, Begossi 
1989a), but families at Buzios only use lizard fat for medicinal purposes. It is 
used to treat jararaca (Bothrops sp.) bites and to cure rheumatism (Table 8 and 
Fig. 4), among other ailments. In northeast Brazil, lizard fat is used against snake- 
bites, asthma, and throat pains (Campos 1967). In 1658, Piso reported that lizard 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



133 




1 




2 M 3 




4 YZM 5 [ 



] 6 EH3NL 





JNE 



<n 




u 




o 

o 
o 

li. 



BEEF 



FISH 



CHICKEN 



TURTLE 



SQUID 



OCTOPUS 



LIZARD 








10 



20 



30 



40 



50 



NUMBER OF INTERVIEWEES 



FIG. 3.— Preference of animal protein: results from interviews (n = 47). Bars are 
coded (1—6) to show the animal protein ranking given by islanders; NL = 



does 



not like it, NE = does not eat it in any way. 




FIG. 4. 



teguixin): this animal 



rheumatism 



134 



BEGOSSI 



Vol. 12, No. 1 



TABLE 8.— Medicinal animals cited in interviews (n = 32). Bold face indicates 
more important usages and animal part used. 



Name 1 



Per- 
centage 
response 



Scientific 

name 



Diseases 



Utilization 



(S) lagarto, 
teiu (lizard) 



81 



Tupinambis teguixin snake-bites 

rheumatism 

asthma 
tetanus 
skin thorns 



The fat (enxundia) is 
drunk with water 
or applied on the 
affected area. Meat 
is eaten for snake 
bites of Bothrops sp. 



(S) tartaruga 
(turtle) 



47 



Chelonia my das 



bronchitis 

asthma 

rheumatism 



The heart is 

toasted and tri- 
tured. The powder 
is drunk with water 
and/or tobacco. For 
rheumatism, the fat 
is massaged on the 
affected area. 



(N) cavalinho 
(sea horse) 



34 



Hippocampus reidi 



(C) chicken 



(N) almofa- 
dinha, barata 
do mar (ray 

eggs) 



22 



13 



Gallus gallus 



Raja cyclophora 



bronchitis 

women's 



Toasted and drunk 
as a tea or with 



hemorrhages pinga (Brazilian 
after childbirth rum) 



cough 

bronchitis 
asthma 



childbirth: 



The fat is massaged 
on chest or drunk 
with water. 



women after Eggs are toasted 



and drunk as tea or 



hemorrhages with pinga. 



(S) caramujo 
(snail) 



(N) peixe 

morcego 

(batfish) 



9 



Megalobulinus sp. 



3 



Ogcocephalus 
vespertilio 



wounds 



bronchitis 



The shell is toasted 
and tritured. The 
powder is applied 
on wounds. 
Toasted meat may 
be used the same 
way. Can be drunk 
with water. 



? 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



135 



Namel 



Per- 
centage 
response 



Scientific 
name 



Diseases 



Utilization 



(S) ouriqo 
do mar, mana- 
caru (sea 
urchin) 



(N) cobra 



3 



3 



Lytechinus 
variegatus 



(?) (snake) 



snake-bites 
(Bothrops sp.) 



It is tritured; the 
green juice is 
drunk. 



against bad The skin is toasted 

luck (mau -olhado) and drunk with 

water. 



(C) jaguariga 
(squirrelfish) 



3 



Holocentrus 
ascensionis 



body wounds The stinger is put 

in contact with 
wounds. 



(N) limo do 
fundo do mar 



3 



algae (?) 



bronchitis 



Used to make 
syrups 



sumed; and (C)— consumed. 



(N)— not consumed; (S)— seldom con- 



fat was used to cure skin wounds. The observation that the most disgusting animal 
for Buzios islanders is also the most important in medicinal terms reinforces the 
argument suggested by Harris (1977), that food taboos may work to " avoid temp- 



consumpt 



will help to maintain 



At Buzios Island, ray and other fish (cavalinho, peixe morcego, and jaguariqd 



medicinals. The medicinal 



(Hippocampus sp.) as an emmenagogue at Buzios Island had already been observed 



Fishermen from the Tocantins River 



many 



exam 



asthma; fat from ;' 



rheumatism 



anim 



medicinal purposes (Table 8) . Camp 



r 



rheumatism in northeast Brazil. Ayres and Ayr 
from Aripuaha (southern Amazon) avoid eating o 
\ due to its medicinal value. 



CONCLUSIONS 



gressive 



)dor, or those classified as carregado, such as camburu (mor 
mackerel and little tunnv), and tinhuna (sargeant major). Ca 



136 



BEGOSSI Vol. 12, No. 1 



avoided during illness; there are other species of fish which are recommended 
for ill persons. These are mostly fish that feed on invertebrates or plankton. The 
baiacu (puff erf ish) is also considered toxic and is not eaten at Buzios. Other 
medicinal animals are subiect to food taboos; the most tabooed is the lizard, which 



medicinal 



adjustments 



may 



The 



some animals as foods may be partially explained by the importance of main 



them for medicinal 



// 



(Begossi 1989b; Begossi and Braga 1992), nature is the "drugstore" of isolated 
human populations; plants (collected and cultivated) as well as animals are used 
for medicinal purposes. As is the case for Harris's (1977, 1985) sacred cow, 
medicinal animals are also too important to be consumed as food. Note that the 
strongest food taboo pertains to the lizard, which is the most frequently used 
medicinal animal (Table 8). According to Rea (1981), a dietary restriction does 
not necessarily protect an animal from human predation, as it may still be taken 
for its feathers, hide, or medicinal/religious value. At Buzios, if medicinal animals 
were also consumed, predation pressure would be stronger on these animals. 
Therefore, the maintenance of a species through food taboos may also have 



purposes 



commu 



about its environment. Islander perception is that some fish are "healthy" 
others "unhealthy." This behavior could be based on an old or a present envi 
mental feature. A taboo may at first be useful. Once in existence, it develop 



may 



conditions ("cultural inertia") (Boyd and Richerson 1985). 



tantly in response to environmental 



observed _ rr w 

depend heavily on fishing have dietary restrictions concerning terrestrial game 
animals. On the other hand, fish is insignificant in the diet of Yanomano, Shuara 
(Jivaro), and Mundurucu populations, and large game restrictions are absent. Ross 
concluded that where aquatic fauna is abundant, aquatic organisms are prized 
as food and land creatures are considered inedible. However, Kiltie (1980) sug- 
gested, based on optimal foraging theory, that food taboos should not depend 
on prey availability, but rather on the ranking of food preferences. 

According to optimal foraging theory (Pyke 1984), when resources are abun- 
dant a predator should specialize in high ranking prey; if resources are scarce 
low ranking prey are included in the diet. If we consider that fishing communities 
such as Buzios Island have plenty of available animal protein, we should expect 
to find the low ranking prey being avoided in these communities (i.e., they will 
specialize in some animal protein items). As suggested by Rea (1981), taboos are 
a luxury: for example, rich riverine peoples with agriculture can afford them but 
peoples from harsh areas have to be more generalist in their food choices. The 
ranking and choice of resources, in association with the other factors already 
discussed, could favor the creation and maintenance of food taboos at Buzios. 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



137 



Three main factors are 



some 



exam 



data). Second, fish that prey on other fish are avoided by unhealthy persons. 
Third, some animals are avoided as food because they are preserved for medicinal 



same factors seem 



behaviors of fishermen along the Tocani 
However, data on fish toxins are much ne( 
factors which affect food taboos in human 



communities from 



Island and Tocantins), India, and Malays 
possible ecological or nutritional factors. 



ACKNOWLEDGEMENTS 



I thank the Conselho Nacional de Desenvolvimento Cientifico e Tecnolo'gico for financial 
support while undertaking doctoral studies at the University of California, Davis, for finan- 
cial aid for the fieldwork at Buzios, and for a current research scholarship and grant; 
the University of California, Davis, for financial support for the Buzios fieldwork; the 
community of Ilhabela, and especially of Buzios Island, for their kind cooperation and 
hospitality; the "Delegacia de Ensino de Caraguatatuba" for giving permission for this 
project at Buzios and offering use of their installations; M.D.O. Campos for introducing 
me to Buzios islanders; P.J. Richerson, B. Orlove, and R. Bettinger (University of California, 
Davis) for helpful comments on an earlier draft of this manuscript; L.F.L. Duarte, Univer- 
sidade Estadual de Campinas, for identifying the molluscs and J.L. Figueiredo, Museu 
de Zoologia, Universidade de Sao Paulo for reviewing the fish identification. Finally, I 
thank L. Junqueira for the French translation of the abstract. 



LITERATURE CITED 



AYRES, JOSE M. and CRISTINA AYRES. 
1979. Aspectos da ca<ja no alto rio Aripu- 
ana. Acta Amazonica 9:287-298. 

BARTHES, ROLAND. 1961. Pour une psy- 
cho-sociologie de 1' alimentation contem- 
poraine. Annales 16:977-986. 

BASSO, ELLEN. 1972. The Kalapalo dietary 
system. Pp. 629-637 in Atti del XL Con- 
gresso Internazionale degli Americanisti. 
Tilgher, Genova. 
— _. 1973. The Kalapalo Indians of 



Central Brazil. Holt, Rinehart & Winston, 
New York. 



— 1978. Comments. Current 

Anthropology 19:16-17. 

BEGOSSI, ALPINA. 1989a. Food Diversity 
and Choice, and Technology in a Brazil- 
ian Fishing Community (Buzios Island, 
Sao Paulo State). Ph.D. dissertation, 
University of California, Davis. Univer- 
sity Microfilms, Ann Arbor. 



_. 1989b. Tabus alimentares na 

Dha dos Buzios, uma comunidade de 
Pescadores. Pp. 253-262 in Anais do III 
Encontro de Ciencias Sociais e o Mar no 
Brasil. Antonio Carlos Diegues (editor). 
Programa de Pesquisa e Conserva^ao de 
Areas Umidas no Brasil. Instituto Ocean- 
ografico da Universidade de Sao Paulo, 
Fundagao Ford, Instituto International 
para a Conservaqao da Natureza, Sao 

Paulo. 

BEGOSSI, ALPINA and FRANCISCO M. de 
S. BRAGA. 1992. Food taboos and folk 
medicine among fishermen from the 
Tocantins River (Brazil). Amazoniana 
12:101-118. 

BOYD, ROBERT and PETER J. RICHERSON. 



Evolutionary 



cess. The 
Chicago. 



138 



BEGOSSI 



Vol. 12, No. 1 



LITERATURE CITED (continued) 



CAMPOS, EDUARDO. 1967. Medicina pop- 
ular do Nordeste. Edigoes O Cruzeiro, 

Rio de Janeiro. 
CANE, SCOTT. 1987. Australian aboriginal 
subsistence in the Western Desert. 
Human Ecology 15:391-434. 



Th 



CARNEIRO 



°gy 



CHAGNON, NAPOLEON and RAYMOND 
B. HAMES. 1980. La "hipotesis pro- 
teica" y la adaptation indigena a la 
cuenca del Amazonas: Una revision cn- 
tica de los datos y de la teoria. Inter- 
ciencia 5:346-358. 

CORREA, IRACEMA F.L. 1981. A Congada 
de Ilhabela na festa de Sao Benedito. 
Editora Livramento, Escola de Folclore, 
Sao Paulo. 

DOUGLAS, MARY. 1969. Purity and 

Danger. Routledge & Kegan Paul, 
London. 

FERRO-LUZZI, GABRIELLA E. 1975. Food 
avoidances of Indian tribes. Anthropos 
70:385-427. 

1980a. Food avoidances of 

pregnant women in Talminad. Pp. 101- 
108 in Food, Ecology and Culture. John 
R.K. Robson (editor). Gordon & Breach, 
New York. 

1980b. Food avoidances dur- 



ing the puerperium and lactation in Tal- 
minad. Pp. 109-118 in Food, Ecology and 
Culture. John R.K. Robson (editor). 
Gordon & Breach, New York. 

FIGUEIREDO, JOSE L. 1977. Manual de 
peixes marinhos do sudeste do Brasil: 
Canoes, raias e quimeras. Museu de 
Zoologia, Universidade de Sao Paulo, 
Sao Paulo. 

and NAERCIO A. MENEZES. 

1978. Manual de peixes marinhos do 
sudeste do Brasil: Teleostei (1). Museu 
de Zoologia, Universidade de Sao Paulo, 
Sao Paulo. 



maim 



hos do sudeste do Brasil: Teleostei (2). 
Museu de Zoologia, Universidade de Sao 
Paulo, Sao Paulo. 

GOPALAKRISHNAKONE, P. 1988. Struc- 
ture of the skin glands of the puffer fish 
[abstract]. Toxicon 26:22. 

GOULDING, MICHAEL. 1981. Man and 
Fisheries on an Amazon frontier. Devel- 
opments in Hydrobioloev, Vol. 4. W. 



HABERMEHL, GERHARD G 
mous Animals and Their 
Springer- Verlag, Berlin. 

HARRIS, MARVIN. 1977. C 
Kings: The Origin of Cul 
Books, New York. 



. 1985. Good to Eat: Riddles of 

Food and Culture. Simon & Schuster, 

New York. 

. 1987a. Comment on Vayda's 



review of Good to Eat: Riddles of Food 
and Culture. Human Ecology 

. 1987b. Food ways 



overview and historical prolegomenon. 

_ .- mm • 



Marvin 



Harris and 



University Press, Philadelphia. 



HASHIMOTO, YOSHIRO. 1979. Marine 
Toxins and Other Bioactive Marine Meta- 
bolites. Japan Scientific Press, Tokyo. 



KILTIE 



ecology 
1-546. 



___, „.. M. and HOKAMA 

YOSHTTSUGI. 1989. Variations in symp- 
tomatology of ciguatera poisoning. Toxi- 
con 27:593-595. 



LANDS. WILLIAM 



Academic 



Orlando. 



LENKO, KAROL. 1965 

caras & Quintais ( 

LEWIS 



para 



meters of a tropical health problem. 
Human Ecology 12:253-274. 
McDONALD, DAVID R. 1977. Food taboos: 
A primitive environmental protection 
agency (South America). Anthropos 72: 

734-748. 
MENEZES, NAERCIO A. and JOSE L. 
FIGUEIREDO. 1980. Manual de peixes 
marinhos do sudeste do BrasU: Teleostei 
(3). Museu de Zoologia, Universidade de 

Sao Paulo, Sao Paulo. 
1985 . Manual de peixes marin- 
hos do sudeste do Brasil: Teleostei (4). 
Museu de Zoologia, Universidade de Sao 
Paulo, Sao Paulo. 



TINDALL, W 



cigu- 



wa - ^— mmmm » ■ • mm ^mmmmy -^ i^ m, mm ^^» m — — - mf 

ateric fish extracts utilizing the guinea 
pig ileum preparation [abstract]. Toxicon 
25:149. 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



139 



LITERATURE CITED (continued) 



MORAN 



The 



system of the Amazonian caboclo. Pp. 
136-159 in Man in the Amazon. C. Wag- 
ley (editor). University of Florida Press, 
Gainesville. 



1977 



Universidade do Rio Grande, Museu 
Oceanografico, Rio Grande. 
ROSENBERG, PHILIP (editor). 1987. Com- 
mon names index: Poisonous animals, 



plants and bacteria. Toxicon 25:799-890. 



vencia: O uso de recursos ao longo da 
rodovia Transamazonica. Acta Amazon- 
ica 7:363-379. 

MOYLE, PETER B. and JOSEPH J. CECH, 
JR. 1982. Fishes: An Introduction to 
Ichthyology. Prentice-Hall, Englewood 
Cliffs, New Jersey. 

MUSSOLINI, GIOCONDA. 1980. Ensaios de 
antropologia indigena e caif ara. Editora 
Paz e Terra, Rio de Janeiro. 

PEREIRA, NUNES. 1974. Panorama da ali- 
mentagao indigena. Livraria Sao Jose, 
Rio de Janeiro. 

PISO, GUILHERME. 1658. Historia natural 
e medica da India Ocidental (reprinted 
in 1957). Departamento de Imprensa 
Nacional, Rio de Janeiro. 

PYKE, GRAHAM H. 1984. Optimal foramn* 



ROSS, ERIC B. 1978. Food taboos, diet, and 
hunting strategy: The adaptation to 
animals in Amazon cultural ecology. 
Current Anthropology 19:1-36. 

RUSSELL, FINDLAY E. 1965. Marine toxins 
and venomous and poisonous marine 
animals. Advances in Marine Biology 
3:255-384. 



MARSHALL 



The 



Chicag 



SMITH, NIGEL J.H. 1976. Utilization of 
game along Brazil's transamazon high- 
way. Acta Amazonica 6:455-466. 

1981. Man, Fishes, and the 



Amazon. Columbia University Press, 
New York. 

TURTON, DAVID. 1978. Comments. Cur- 
rent Anthropology 19:26-27. 
theory: A critical review. Annual Review VAYDA, ANDREW P. 1987. Explaining what 



of Ecology and Systematics 15:523-575. 

QUOD, J.P. and ANNE-MARIE LEGRAND. 
1988. Effects of partially purified cigua- 
toxin from moray-eel, Gymnothorax java- 
nicus, on action potential of isolated rat 
heart [abstract]. Toxicon 26:36. 

RE A, AMADEO M. 1981. Resource utiliza- 
tion and food taboos of Sonoran desert 
peoples. Journal of Ethnobiology 1:69-83. 

REICHEL-DOLMATOFF, GERARDO. 1976. 
Cosmology as ecological analysis: A view 
from the rain forest. Man 11:307-318. 

RIBEIRO. BERTA G. 1987 O inrlin na ml- 



people eat; A review article. Human 
Ecology 15:493-510. 

VON BRANDT, ANDRES. 1984. Fish Catch- 
ing Methods of the World. Fishing New 
Books, Farham, England. 

WATABE, S., Y. SATO, M. NAKAYA, N. 
NOGAWA, K. OOKASHI, T. NOGU- 
CHI, N. MORIKAWA, and K. HASHI- 
MOTO. 1987. Distribution of tritiated 
tetrodoxin administered intraperitoneally 
to pufferfish. Toxicon 25:1283-1289. 

WILLEMS, EMILIO. 1952. Buzios Island. 
University of Washington Press, Seattle. 



tura brasileira. Organizag ao das Nacoes WILSON, CHRISTINE F. 1980. Food taboos 

XT • 1 _ - m A _. ------ -._- ■ 



Unidas para a Educagao, Ciencia, e Cul- 
tura, Rio de Janeiro. 
RIOS : E.C. 1985. Seashells of Brazil. Funda- 
£ao Cidade do Rio Grande, Fundagao 



at childbirth: the Malay example. Pp. 
67-74 in Food, ecology and culture. 
John R.K. Robson (editor). Gordon & 
Breach, New York. 



BOOK REVIEW 



Plantas y el Hombre: Memorias del Primer Simp 
botanica y Botanica Economica. Montserrat Ri 
Pederson (editors). Quito, Ecuador: Ediciones AB' 
437. No price given (paperback). ISBN 9978-99-C 



Henrick Borgtoff 



140 



Vol. 12, No. 1 



lament the difficulty of finding out what 



conferences they have not attended; even when published, the proceedings are 
often out-of-date by the time they see the printed page. Rios and Pederson are 
to be congratulated for their efficiency in issuing this compendium of papers 
presented at the First Ecuadorian Symposium of Ethnobotany and Economic 
Botany, held in Quito from 27 February-2 March, 1990. 

The volume contains papers written by 45 contributors representing eight 
countries. Of a total of 33 articles, five are written in English and 28 in Spanish; 
abstracts are given in both languages. The papers are divided among three 
sections: (1) conservation and management of useful plants; (2) medicinal plants; 
and (3) ethnobotany and its application. 

As with most symposium proceedings, the range of topics is diverse and the 
quality of the papers uneven. Many are based on research carried out in Ecuador, 
but some contributions describe projects in Mexico, Colombia, and other Latin 

American Countries. Others ranc*P fiirthpr afiplH- tVip mrlncinn of a rtanvr nn thp 



especially 



birth control plants to subdue 



most 



number of papers are broad 



primary plant genetic resources of Ecuador and discusses 



program for their conservation 



informative summary 



wild palms. Estrella, initiator of the symposium, describes 
nts in use before the arrival of the Spanish. A well- written 
tie history of ethnobotany and describes its trajectory. 



summaries 



amtry-wide inventories of medicinal plants in Mexico t 
Honduras (House) and comparative ethnobotany in the northern 

A.), among others. One hopes that these "tip 



Mexico (Martinez 



treatments 



The 22-page bibliography is particularly valuable for ethnobotanists who work 
in South America; many articles of limited distribution are included. Indices of 
scientific and vernacular names are useful for readers interested in folk classifica- 
tion or in a particular taxonomic group. 

The book was copublished by the Herbario QCA, Departamento de Cie'ncias 
Biologicas, Pontificia Universidad Catolica del Ecuador, Quito, Ecuador. Inquiries 
should be addressed to the editors at this address. 



Gary J. Martin 
Anthropology Department 
University of California 
Berkeley, CA 94720 



/. Ethnobiol. 12(1): 141-152 



Summer 1992 



SHORT COMMUNICATION 



SOME OBSERVATIONS ON THE USE OF 

MEDICINAL PLANTS FROM PRIMARY AND SECONDARY 

GROWTH BY THE RUNA OF EASTERN LOWLAND ECUADOR 



EDUARDO O. KOHN 

32 Sturges Way 
Princeton, NJ 08540 



Amazonian rain forests are often depicted in the popular literature as "drug 
stores/' with the implication that indigenous peoples use them much like 
city- dwellers go to the corner pharmacy for an aspirin or some other remedy. 
Recent observations I made on the medical culture in a Runa (Quichua-speaking 
Indian) community in eastern lowland Ecuador challenge this notion. Indeed, 
numerous primary forest taxa are recognized medicinals but few receive frequent 
use. An examination of medicinal plant usage patterns revealed secondary vegeta- 
tion as a much richer source of useful species. 

Observations on medicinal plant use were made from February to July 1990 
in Rio Blanco (Napo province, 77°40'W; 1°00 / S, 440 m elevation), a small Runa 
settlement (population ca. 100). Rio Blanco residents hunt, fish, and grow crops 
such as manioc, and peach palm (Bactris gasipaes H.B.K., Arecaceae). Panned gold, 
coffee, cacao, and corn are sold in order to purchase items such as salt, gun 
powder, machetes, clothing, and Western medicines. Despite the availability of 
Western goods including remedies, the Runa continue to rely heavily on medicinal 
plants. Runa medical culture is examined in greater depth elsewhere (Kohn 1992). 

In order to determine usage patterns, medicinal plants were classified 
according to the habitat in which they were collected: primary forest (rucu sacha), 
secondary forest (mauca), and garden (chagra). Most secondary forest plots in 



much 



when the area was first settled. Primary 



in 



The classification of medicinal plants by habitat confirmed an expected parallel 
and revealed an unanticipated paradox. Although medicinal plant occurrence 
paralleled the general trend of increased species diversity during forest succes- 
sion, the importance of a plant assemblage as a source of remedies was not related 
to the number of useful species within it, indicating that other factors determine 
usage patterns. Fifty-four percent of the 191 medicinal species collected (Table 1) 
occurred in primary forest, 29% were growing in secondary forest, and 17% 
occurred in gardens (Fig. 1). However, medicinals from secondary forest and 
gardens were more frequently used than those from primary forest. I observed 
applications with 25 different taxa: 40% of these were collected in gardens, 
36% in secondary forest, and 24% came from primary forest (Fig. 2). Thirty-one 



142 



KOHN 



Vol. 12, No. 1 



17.00% 



29.00% 




54 . 00% 



Prim. Forest 
Sec. Forest 
Garden 



FIG. 1.— The distribution of the collected me 



were 



percent of the 32 garden medicinals collected and 16% of the 56 secondary forest 
medicinals were observed being used. Only 6% of the 103 primary forest taxa 
were observed in use (Fig. 3). 

Although medicinals appear to be incorporated into the Runa materia medica 



because of efficaciousness— Maries 
chemical literature 
collected among tl 



(1988) found pharmacological 



other considerations encourage 
disproportionate reliance on plants from disturbed habitats. Plant conspicu- 
ousness, cultural taboos, multipurpose utility, and ease of transplantation may 
encourage the use of some medicinals instead of others. 

Early successional species tend to be gregarious and thus more conspicuous 
and accessible than the often highly dispersed forest taxa. Furthermore, although 
primary forest occurs no more than 200 m from any doorstep, the Runa do not 
venture lightly into the forest— the home of potentially dangerous spirits (supai) 



accomplish specific tasks. Women 



the forest alone. 

Frequently used plants are also cultivated in gardens and other disturbed 
and many have multiple uses. For example, a variety of preparations c 
cultigen tahucu (Nicotiana tabacum L„ Solanaceae) are used medicinally: 
smoke is blown over patients during shamanistic curing sessions; moistened 1 
are administered as an emetic: tobacco iuirp is asnirateH nacaih, to n-oaf h^aH 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



143 



24.00°/« 



40.0% 




Prim. Forest 
Sec. Forest 
Garden 



36.00% 



FIG - 2.— The distribution of medicinal 
according to the habitat in which thev 



and drunk as a narcotic; tobacco resin is used to expel subcutaneous hualun curu 
(probably Dermatobia hominis) larvae; and poultices are applied to treat a variety 
of ailments ranging from chest pains to infected blisters. 

For convenience, frequently utilized forest medicinals are often transplanted 
to disturbed areas. Of the 34 species recorded in cultivation, eight had recently 

een transplanted from primary forest. Some transplants, such as dunduma 
(Cyperus sp., Cyperaceae), a jus huasca (Mansoa cf. alliacea (Lam.) A. Gentry, 

ignoniaceae), antarun caspi (Ouratea sp., Ochnaceae), yahuar purungu panga 
(Mikania sp., Asteraceae), and tahucu sisa (Campanulaceae), are cultivated in open 
gardens. Others, such as machacui huishu (Malpighia cf. glabra, Malpighiaceae), 
are planted beneath secondary growth or in plantings of coffee intercropped with 
cacao and peach palm, presumably because these associations duplicate late 
successional habitats from which the taxa were transplanted. Still other trans- 
plants, such as maticara (Clavija sp., Theophrastaceae) and curarina (Potalia amara 

ublet, Loganiaceae), are more broadly tolerant and persist as resources in 
allowed areas. However, not all primary forest medicinals are easily transplanted. 
Blanco residents have been unable to cultivate the commonly used midcanopy 
reechucchu huasha (Maytenus sp., Celastraceae) apparently because, like many 
other climax taxa, it is drought sensitive. Furthermore, slow-growing late succes- 
sional plants may not produce useful quantities of secondary metabolites for 
several decades, making cultivation impractical. 



144 



KOHN 



Vol. 12, No. 1 



40 



CD 

CD 




O 

</) 
CO 

CD 

J) 



CD 

o 



5 



CO 

Cl 

o 

CD 
D> 

CO 



CD 

o 

CD 

Q. 



30 



20 



10 








Prim. Forest Sec. Forest 



Garden 



Habitat 



observed 



The percentage of plants from each habitat whose 



The high species diversity of primary forest and the many medicinals found 
there have encouraged observers to overemphasize climax taxa and to underrate 
the importance of secondary communities to native pharmacopoeias. Primary 
forest is certainly an important resource but more so as a source of medicinal 
propagules than as a "drug store." Surveys elsewhere would determine whether 
the pattern of plant usage in Rio Blanco is a general one among Amazonians. 



ACKNOWLEDGEMENTS 



kindness 



W 



J 



Missouri 



Quit 



x VA MM A Tx M ^tw ^uaiuiwiiu ue ^lencias i\aiurau:& m *>*•** ~- 

I am indebted to the botanists at both of these institutions, in particular David Neill, for 
determining these collections. I thank David Benzing for reviewing earlier drafts of this 
manuscript. 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



145 



TABLE 1.— Medicinal plants collected in Rio Blanco. The common Quichua 
Spanish name is given after the Latin name. This information is followed 1 



which 



or the malady foi 
the plant in use. 



Scientific 
Name 



Common 
Name 



Habitat 



Use 



Abuta grandi folia 



Acacia glomerosa 
Acalypha sp. 



Allophylus sp. 

Amaryllidaceae 

Annonaceae 



Annonaceae 



Anthurium ernestii 
A. pendulifolium 



A. polyschistum 



Anthurium sect., 
Pteromischum sp. nov 



Anthurium 



sp 



Anthurium sp. 



Aristolochia sp. 
Asteraceae 
Bactris sp. 



Banisteriopsis caapi 
Batocarpus sp. ? 
Batocarpus sp. ? 

Bauhinia sp. 

Begonia parviflora 
Begonia sp. 

Bixa orellana 

Blechum browneii 
Bolbitis sp. 

Brosimum utile 
Brownea sp. 

Brugmansia sp 
Brugmansia sp 



? 



payanchi 

yurac, 

yahuate 

caspi 

huaranga 

ita 

quhuilchic 

cambia 

sacha cebolla Garden 

lichihua 



Secondary forest diarrhea, regulate 



menstruation 



Primary forest colds 

Secondary forest diarrhea, bloody 



Primary forest 



Primary forest 



diarrhea 
snake bite 
tumors 
cataracts 



Primary forest 



yurac 

pechiche 

remo caspi, 

pinja cara 

zingra panga Primary forest 

suni zingra Primary forest 

panga 

not given 



body pains, diarrhea, 
stomach ache 
arthritis, rheumatism 
arthritis, rheumatism 



Primary forest 



not given 



Primary forest 



pasumu 
panga, 
concordia 
yami panga 



Primary forest 



topical parasitic and fungal 

infections 

love magic 

postpartum inflammations, 
pains, malaise 



Primary forest 



with 



Primary forest 



Primary forest 



saragosa 

aya huachi 

ucu mari 

chunda 

aya huasca 

negro caspi 

sacha 

paparahua 

yacu yutsuc 

bibitsic 

auru panga, 

pungi panga 

manduru 

quibiu quihua Garden 



Secondary forest acne, hypopigmentation 



sore joints 



Secondary forest hallucinogen 

Primary forest 
Primary forest 



tonic 

infected wounds 



Primary forest 
Secondary forest 
Primary forest 



diarrhea 
inflammations 

medicinal 



Secondary forest 



cutu chupa 



Secondary forest 



pucuna caspi Primary forest 

Primary forest 



cruz caspi 
huantuc 
sasi huantuc 



Garden 
Garden 



accelerate birth 

sprains 

diarrhea 

vermifuge 

regulate menstruation 

body pains, inflammations 

hallucinogen, body pains 



Use 
Ob. 



Yes 



Yes 

No 



No 
No 
No 



No 



No 
No 



No 



No 



No 



applied to the legs of children No 



diarrhea, cough, stomach ache No 



No 
No 



Yes 

No 
No 



No 
No 
No 



No 

Yes 
No 

No 
No 
No 
No 



146 



KOHN 



Vol. 12, No. 1 



TABLE 1. (continued) 



Scientific 
Name 



Common 
Name 



Habitat 



Use 



Brunfelsia cf. 
grandiflora 



B. grandiflora 



Brunfelsia sp. 



Caladium sp 

Calliandra 

angustifolia 



Calyptranthes sp. 
Campanulaceae 
Campelia zanonia 
Capsicum sp. 



Carludovica palmata 
Cephaelis sp. or 
Psychotria sp. 
Cephaelis williamsii 



Citrus sp. 
Clavija sp. 
Clavija sp. 



Clavija sp. 



Clidemia sp. 



Columnea sp. 



Commelinaceae 



Cordia nodosa 



Costus sp. 
Coussarea sp. 



Croton lechleri 



Cucurbitaceae 
Cuphea sp. 



Cyathea sp. 



sacha chiri 
huayusa 



Primary forest 



chiri 
huayusa 



Garden 



sacha chiri 
huayusa 



Primary forest 



postpartum fevers, body 
pains, emetic, narcotic, 
hunting luck, colds 
postpartum fevers, body 
pains, emetic, narcotic, 
hunting luck, colds 
postpartum fevers, body 
pains, emetic, narcotic, 
hunting luck, colds 



hualun mandi Secondary forest 
ichilla yutsu, Primary forest 



larvae 



diarrhea 



chiparo 
pequeno 
sani mulchi 
tahuco sisa 



Primary forest 
Garden 



quilun quilun Secondary forest 



uchu 



Garden 



menstruation, hemorrhage 

liver ailments, tumors 

cuts 

expel illness-bringing spirits 

from the body and house, 

diarrhea 



lisan 

chiri panga 



Secondary forest "zingra" (rheumatism?) 



Primary forest 



yan chipi 
panga 
limun 
mati cara 
llushti mati 
cara 

jatun mati 
cara 



Primary forest 



rattle/ fan for curing 

ceremonies 

colds 



Garden 
Garden 
Secondary forest 



diarrhea, nausea, fever 
snake bite 
snake bite 



Primary forest 



snake bite 



Clidemia heterophylla uchan 



Primary forest 



body pains, stomach ache, 



colds 



uch 



an 



Primary forest 



body pains, stomach ache, 



colds 



dumbiqui 
callu panga 

shungu nanai Primary forest 
panga, supi 
panga 
arana caspi 



Secondary forest hemorrhages, reduce 



menstrual flow 
liver 



Primary forest 



sacha iru 



Primary forest 
pungi panga Primary forest 
yurac 



snake and spider bites, 

infections 

diabetes 

gas 



Ian iqui, 
sangre de 
drago 
chia 



Secondary forest cuts, anemia, kidney, 

stomach 



Secondary forest scabies 



ichilla panga Primary forest 



arthritis 



shica 
pichichi 



Primary forest 



Cyclanthus bipartitus tsicta, papancu Primary forest 



cuts, mumps? 
snake bites, cataracts 



Use 
Ob. 



No 



No 



No 



No 
No 



No 
No 
No 
Yes 



No 

Yes 



No 



Yes 

No 
No 



No 



No 



No 



No 



No 



No 



No 
No 



Yes 



No 
No 



No 
No 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



147 



Scientific 


Common 






Use 


Name 


Name 


Habitat 


Use 


Ob. 


Cymbopogon citratus 


hierba luisa 


Garden 


stomach ache, headache, 
body pains 


No 


Cyperus sp. 


dunduma 


Garden 


diarrhea 


No 


Desmondium 


llutari 


Secondary forest 


hemorrhages, menstrual 


No 


axillare 


quihua 

batan 

quihua 




irregularities 




Desmoncus sp. 


bara casha 


Primary forest 


tumors 


No 


Dicranopygium sp. 


i 

yucu lisan 


Primary forest 


arthritis, rheumatism 


No 


Dieffenbachia cf. 


chaha lalo 


Primary forest 


luck 


No 


daguense 










Dieffenbachia sp. 


chaha lalo 


Primary forest 


hunting luck 


No 


Diplopterys cabrerana 


chali panga 
huasca 


Secondary forest 


hallucinogen 


No 


Dracontium cf. 


machacui 


Primary forest 


snake bite 


No 


loretense 


mandi 


9f 






Dracontium sp. 


machacui 
mandi 


Secondary forest 


snake bite 


No 


Eschweilera sp. 


cushillu 
manga cara 


Primary forest 


tuberculosis 


No 


Eugenia cf. 


yana muyu 


Primary forest 


cavities, menstrual 


No 


subterminalis 


mulchi 




irregularities, diarrhea 




Ficus sp. 


plor ila 


Primary forest 


diarrhea 


No 


Fictts sp. 


puca ila, 
higueron 


Primary forest 


vermifuge 


No 


Fittonia verschaffeltii 


nina curu 
panga 

yahuar panga 


Primary forest 


"caracha" cutaneous 
parasites 


No 


Fittonia sp. 


yahuar 
panga 


Primary forest 


fishing luck 


No 


Geonoma sp. 


macana 
panea 


Primary forest 


abdominal pains from 
overexertion 


No 


Gesneriaceae 


gallu panga, 
yahuar sisa 
panga 


Primary forest 


hemorrhages, snake bites 


No 


Gesneriaceae 


gallu sisa 
quihua, puca 


Primary forest 


menstruation 


No 


Gesneriaceae 


gallu sisa 


Primary forest 


menstruation 


No 


Gesneriaceae 


inda paju 


Secondary forest 


infection 


No 


Grids neuberthii 


panga 
piton 


Primary forest 


emetic, malaria, hunting 
luck, postpartum 
abdominal pains 


No 


Guadua sp. 


huama 


Secondary forest 


inflammations 


No 


Gurania sp. 


sapallu 
panga, 
ahuas panga 


Primary forest 


scabies 


No 


Heliconia 


lliquiri siqui 


Secondary forest 


accelerate birth 


No 


aemygdiana 


panga 






*. ■ 


Heliocarpus 


damua 


Secondary forest 


accelerate birth 


No 


americanus 











148 



KOHN 



Vol. 12, No. 1 



TABLE 1. (continued) 



Scientific 
Name 



Common 
Name 



Habitat 



Use 



Herrania sp. 
Hyptis pectinata 



Garden 



tuberculosis 



Hyptis sp. 



Secondary forest cough, fever, contraceptive 



Ilex guayusa 



sacha cambia Secondary forest snake bite 

caballu 

quihua 

chiquis 

quihua, 

chagras 

quihua, 

albahaca 

huayusa 



Garden 



stimulant, mouth rinse, 
pains, fever, labor pains, 
mixed with hallucinogen 

Banisteriopsis caapi 



Inga edulis 
Mar tea deltoidea 
Iryanthera sp. 



pacai, guaba Secondary forest colds 



pushiua 



Primary forest 



sicu huapa Primary forest 



Jacaranda glabra 



Lauraceae 
Leonia glycycarpa 



cupa panga 
yurac 
anis ahua 



Libadeum sp. 



Primary forest 

Primary forest 
tamia muyu Primary forest 
yurac 

yacu maria Primary forest 
panga 
chunda 
huasca 



warts 

diarrhea, to stop 

menstruation 

scabies 



mumps 



arthritis 



Primary forest 



abortive, hasten menstru- 
ation 



Lomariopsis 
japurensis 

Lonchocarpus nicou timun ambi Secondary forest ichthyotoxin, expel spirits 

poison to commit suicide 
Macfadyena uncata tuta pishcu Secondary forest hunting and love magic 



Maieta guianensis 



sillu 
uchan 



Primary forest 



Malpighia cf. glabra machacui 

huishu 
escobilla 



Garden 



body pains, stomach ache, 

colds 
snake bite 



Malvaceae 



Garden 



headache 



Mansoa cf . alliacea ajus huasca Secondary forest colds, body pains, fevers, 



Mayna odorata 



measles 
ichilla mati Secondary forest snake bites 



cara 



Maytenus sp. 



chucchu 
huasha 



Primary forest 



stomach ache, diarrhea, 
body pains, anemia 



Melastomataceae puca paytsic Secondary forest skin infections 



Miconia sp. 
Mikania sp. 



panga 

ahua paytsic Primary forest 



Mollinedia sp. 



yahuar 
purungu 
panga 
urcu chiri 
huayusa 
urti tullu 



Garden 



diarrhea 

snake bites, cuts, 

menstruation 



Primary forest 

Monolena sp. urti tullu Primary forest 

Monstera sprueana iqui quihua, Primary forest 

suelda con 

suelda 



medicinal 



colds 

anemia, inflammations 



Use 
Ob. 



No 
No 



No 



Yes 



No 
No 
No 



No 



colds, cough, stomach ache No 



No 



No 



No 



Yes 



No 



No 



No 



No 

No 



No 



Yes 



Yes 



No 

No 



No 



No 
No 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



149 



Scientific 
Name 



Common 
Name 



Habitat 



Use 



Moraceae 

Myroxylon balsamum balsamo 

yurac 
Nicotiana tabacum tahucu 



chinchi yurac Primary forest 



tumors 



Orthostichopsis 
tortipilis 



Otoba parvifolia 
Ouratea sp. 



Pariana sp. 

Parkia multijuga 
Pedaliaceae 



Petiveria alliacea 



Philodendron sp. 
nov. ? 



P. cf. uleanum 



Philodendron sp. 



nov. 



Philodendron sp. 
possibly nov. 
Phytolacca cf. 
rivinoides 
Piper sp. 

Piper sp. 

Piper sp. 
Piper sp. 

Piper sp. 
Piper sp. 
Piptadenia sp. 



Piptadenia sp. 
Plagiochila sp. 



Secondary forest diarrhea, colds 



Garden 



tuca 



Primary forest 



huapa yura Primary forest 
iqui 

ichilla cara Primary forest 
amarun caspi 



ritual cleansing, sore 
muscles, head colds, sore 
throats, sinus congestion, 
narcotic, colic, body pains, 
tonic, "shungu rigushca 
paju," bot fly larva, protec- 
tion from snake venom 
cuts, internal pains, 
stomach ache, avoid being 
poisoned by snake venom 
diarrhea 



zuru panga 



Garden 



sutanga 



Primary forest 



virgin mama Garden 

sisa 

condiciun Garden 

panga 

sacha lalo 



diarrhea, postpartum 
abdominal pains, 
menorrhagia 
rattle/fan for curing 
ceremonies 
hypopigmentation 

accelerate birth 



colds, cough 



Primary forest 



nan ambi 

huasca 

lalo 



Primary forest 
Primary forest 



malaise, inflammations, 
internal body pains, 
hunting luck 

snake bite 



hunting luck 



ichilla 
zingra panga 

huataracu 
muyu 

shia, armallu 
ringri panga 
Santa Mana 
de anis 
urcu shia 



Primary forest 



rheumatatism, arthritis 



Secondary forest tuberculosis 



Primary forest 



diarrhea, nausea 



Garden 



colds, body pains 



Primary forest 



diarrhea, nausea 



mucu tullu Secondary forest diabetes 
yurac 

chugri yuyu Primary forest 
basu panga Primary forest 
chunda rucu Primary forest 
paju yurac, 
urcu tamburu 



cuts 

hernia 

blisters 



asna 



huaranga 



Secondary forest diarrhea 



tuca 



Primary forest 



cuts, internal pains, 
stomach ache, avoid being 
poisoned by snake venom 



Use 

Ob. 



No 
No 



Yes 



No 



No 



No 



No 



No 
No 



No 



No 



No 



No 



No 



No 



No 



No 



No 

No 



No 
No 
No 



Yes 



No 



150 



KOHN 



Vol. 12, No. 1 



TABLE 1. (continued) 



Scientific 
Name 



Common 
Name 



Habitat 



Use 



Use 
Ob. 



Poaceae 



shinglu 



Polybotrya 
crassirhizoma 
Portulaca sp. 



Secondary forest rheumatism, whooping 

cough 



No 



cutu chupa Secondary forest diarrhea 



No 



Garden 



Potalia amara 

Pothomorphe 

peltata 

P. umbellata 



chucli 

quihua 

curarina 

cari maria 

panga 

maria panga Garden 



kidney, diabetes 



No 



Secondary forest snake bites 



Garden 



No 

No 



Yes 



Pouteria caimito 



abiyu 



Garden 



No 



Protium nodulusum siri quillu 
Psychotria viridis 



sami ruca 



Primary forest 
Garden 



Renealmia sp. 
Renealmia sp. ? 



ichilla 

shiguango 
jatun 

shiguango 



Secondary forest snake bite 
Secondary forest snake bite 



tooth ache, colds, 

inflammations 

inflammations, 

tooth ache, colds 

rattle/fan for curing 

ceremonies, scabies 

colds 

mixed with the hallucinogen No 

Banisteriopsis caapi 

No 



No 



No 



Retiniphyllum sp.? urcu matiri Primary forest 



Rheedia 



sp 



Rheedia sp. 



pungara 
chunda rucu 
mulchi 



Primary forest 
Primary forest 



rheumatism, hunting luck, 

colic 

scabies, "charas" 

postpartum colic 



No 



No 
No 



Ricinus communis 
Rubiaceae 



tonic 



Salpichlaena 

volubilis 

Scleria sp. 

Selaginella 

exaltata 

S. geniculata 



atalpusamuyu Secondary forest arthritis 

mucu caspi Primary forrest 

yurac 

urcu tutayu Primary forest 



No 
No 



diarrhea 



Yes 



Serjania sp. 



shinglu 

hualumbu 
huasca 
zancudo 
quihua 



Secondary forest colds 



Primary forest 
Primary forest 



colds, cough 
insect repellent 



No 
No 



No 



chunda rucu Secondary forest "chunda rucu paju" blisters Yes 



Siparuna sp. 
Siparuna sp. 



Smilax sp. 
Smilax sp. 



Solanaceae 



paju huasca, 
rayu huasca 
malagri 
panga 
mal agri 
panga 

quilambu 
casha 

quilambu 
casha hausca 



Secondary forest infections 
Secondary forest infections 



No 



No 



primary forest 
Primary forest 



tumors 



No 



cataracts, diarrhea 



No 



Solanaceae 



asna panga Primary forest 

yurac, ataqui 

panga 

pupu huasca Primary forest 



epilepsy 



No 



Solanaceae 



avoid bleeding from 
umbilical chord 



No 



apumpu cara Secondary forest measles 



No 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



151 



Scientific 


Common 






Use 


Name 


Name 


Habitat 


Use 


Ob. 


Solanaceae 


yacu huantuc 


Primary forest 


medicinal 


No 


Solanaceae 


yacu caspi 


Secondary forest 


headache 


No 


Solanaceae? 


ilia huanga 
butui 


Secondary forest 


tuberculosis 


No 


Solarium sp. 


ataqui panga, 
asna panga 


Primary forest 


epilepsy 


No 


Solarium sp. 


nina curu 
paju panga 


Secondary forest 


"nina curu paju" (a para- 
sitic skin infection) 


Yes 


Solarium sp. 


inda paju 
panga 


Secondary forest 


"ituc paju" (a parasitic 
skin infection) 


No 


Sparattanthelium 


dunduma 


Secondary forest 


stomach ache, diarrhea, 


Yes 


glabrum 


huasca 




malaria, body pains 




Spathiphyllum sp. 


hualun 
mandi 


Primary forest 


expel bot fly larvae 


No 


Spigelia sp. 


cuica quihua 


Garden 


vermifuge 


Yes 


Swartzia simplex 


negro caspi 


Primary forest 


tonic 


No 


Tabernaemon tana 


tsicta 


Primary forest 


labor pains 


No 


sananho 










Tabernaemontana 
sp. 

Thelypteris 


uchu tsicta 


Primary forest 


skin irritations 


No 


suni panga 


Primary forest 


arthritis 


No 


angustifolia 


shica 








Theobroma bicolor 


patas 


Secondary forest 


diarrhea 


Yes 


T. cacao 


cacao 


Secondary forest 


hypopigmentation 


No 


Tococa sp. 


uchan 


Primary forest 


medicinal 


No 


Tournefortia sp. 


tsaca huasca, 
sacha purutu 
huasca 


Garden 


infections 


No 


Tovomitopsis 


uhu angu 


Primary forest 


colds 


No 


membranacea 


yurac 








Tovomitopsis sp. ? 


uhu angu 


Primary forest 


colds 


No 


tirera lacinata 


chunda chini 


Primary forest 


body pains, measles 


Yes 


Urera sp. 


papaya chini 


Garden 


body pains, stomach ache 


Yes 


Urticaceae 


biu chini 


Secondary forest 


"huairashca," measles 


No 


Valerianaceae 


tucsi quihua, 
api quihua 


** 

Garden 


sharp internal pain 


No 


Verbenaceae 


virvina 


Garden 


malaria, diarrhea, fever 


No 


Vernonia patens 


liunchic, 
chilca 


Secondary forest 


eye irritation, diarrhea, 
fever, bronchitis 


No 


Weigeltia sp. 


sacha tahucu 


Primary forest 


body pains, headache, 
malaise, hunting luck 


Yes 


Witheringia 
solanacea 


zimbiyu 
yurac panga 


Secondary forest 


malaria, diarrhea, 
inflammations, scabies, 


No 


Zingiber officinale 


ajiringri 


Garden 


diarrhea, nausea, stomach 
ache 


Yes 


Undetermined 


nina curu 
paju yurac 


Secondary forest 


"nina curu paju" (a para- 
sitic skin infection) 




Undetermined 


nina curu 
panga 


Secondary forest 


eye infections 


No 



152 



KOHN 



Vol. 12, No. 1 



LITERATURE CITED 



KOHN, EDUARDO O. 1992. La cultura 
medica de los Runas de la region Ama- 
zonica Ecuatoriana. Ediciones Abya- 

Yala, Quito. 
MARLES, ROBIN J., DAVID A. NEILL, and 
NORMAN R. FARNSWORTH. 1988. A 



contribution to the ethnopharmacology 
of the lowland Quichua people of Ama- 
zonian Ecuador. Revista de la Academia 
Colombiana de Ciencias Exactas, Fisicas, 
y Naturales 63(16):111-120. 



/. Ethnobiol. 12(1): 153- 156 



Summer 1992 



SHORT COMMUNICATION 

USE OF CINDER CONK (Inonotus obliquus) BY THE GITKSAN 
OF NORTHWESTERN BRITISH COLUMBIA, CANADA 



LESLIE M. JOHNSON GOTTESFELD 

Department of Anthropology 

University of Alberta 
Edmonton, Alberta T6G 2H4 



The Gitksan Indians of northwestern British Columbia, Canada, gathered the 
"sterile conks" of the cinder conk fungus, Inonotus obliquus (Pers.: Fr.) Pilat 
(Hymenochaetaceae) (Holsten et al. 1985; Gilbertson and Ryvarden 1986). These 
massive black eruptions of fungal tissue externally resemble charred cork; under 
the dark crust they are a yellowish-brown color and have a corky texture. They 
emerge from cankers on living birch trees (Betula papyrifera Marsh.) (Fig. 1 and 
Fig. 2). The crusty sterile conk was broken off the trunk for use. Cinder conk had 




FIG. 1. 

stem, Cedarvale, British Columbia, Canada 



sterile conk growing on living Betula papyrifera Marsh 



154 



GOTTESFELD 



Vol. 12, No. 1 




FIG. 2.—Inonotus obliquus sterile conk in section showing yellowish brown intenor 
and black surface. Collected from living Betula papyrifera stem, Quesnel, British 
Columbia, Canada. 



two principal uses: for moxibustion treatment of swollen athritic joints, and as 
tinder or a slow match for making and transporting fire. Use of cinder conk tor 
moxibustion continues; use for a slow match has been supplanted by matches 
and other modern means of kindling fire. 

After conversations with a Gitksan elder who described the use of this unusual 
fungal structure, I collected specimens of cinder conk and brought a collection 
of the fungus to the elder to verify the identification. A voucher (Gottesfeld 
Eth. 54, ALT A) is deposited with the University of Alberta Herbarium. 

The Gitksan have two words for cinder conk: mii'hlw and tiiuxw. I am uncer- 
tain if these words have distinct meanings; they may represent intervillage dialect 
variants with the same meaning. A Gitksan elder describes cinder conk and its 



Summer 1992 



JOURNAL OF ETHNOBIOLOGY 



155 



medicinal use as follows: "Mii 



the black growth from the crack in the birch 
tree. Like yellow cotton inside. If you cut it off, use the yellow cotton stuff. Take 
a sliver like a match stick and burn it for pain in joint." (interview notes, 2/20/87). 
According to the elder, after the sliver of cinder conk was burned to the skin on 



me dnectea joinr, a special salve was then applied to the burn wounds. This treat- 
ment was reported to be effective in reducing the swelling, and presumably the 
discomfort, of the joint. Another Gitksan informant reported that his grandmother 
applied pitch after burning cinder conk slivers to treat her arthritic knee. 

A handful of cinder conk was used by the Gitksan as a slow match. It was 
carried as an ember to kindle a new fire. "Bring fire from the morning camp 
fifteen or twenty miles. Wrap in birch bark. Light fire in the evening. It remains 
an ember for a long time. It doesn't smoke." (interview notes, 2/20/87). 

Three other Indian groups of northern and northwestern North America also 
named and used cinder conk. The Wet'suwet'en of northwest British Columbia 
have two names for cinder conk (dicic'ah ci'ists'o' and tl'eyhtse). The identifi- 
cation and names were verified by showing pieces of a sterile conk to an elder 
from Hagwilget Village (interview notes, 6/14/89). She described a moxibustion 
treatment using cinder conk similar to that used by the neighboring Gitksan. The 
Wet'suwet'en also used it as a slow match. This use is recalled in a traditional 



start a fire. 



ember 



According to Kari (1987), the Tenaina of southcentral Alaska used cinder conk, 
which they call k' atnitsayi, for a punk or tinder with a fire drill, as a slow match, 



treatment 



taxonomic 



the identity of the fungus. 



descript 



Woods 
fomen tarius 



arthritis. In addition, J. obliquus was used to make a tea. 



ACKNOWLEDGEMENTS 



Thanks are due to David Green, Charles Austin, Marianne Austin, and David Harris 
for sharing information on the names and uses of cinder conk. I thank Dr. Randy Currah 
for verifying my identification and Brenda Callan of the Pacific Foresty Centre in Victoria, 
B.C. for providing the photograph for Fig. 2. The manuscript was improved by reviews 
by Nancy Turner, Brenda Callan, and Duncan Morrison. 



LITERATURE CITED 



GILBERTSON, ROBERT L., and LEIF 
RYVARDEN. 1986. North American 
Polypores, Fungiflora, Oslo. 2 vols. 

HOLSTEN, EDWARD H., PAUL E. HEN- 
NON, and RICHARD A. WERNER. 
1985. Insects and diseases of Alaskan 
forests. USDA Forest Service, Forest 



Pest Management, State and Private 
Forestry, Alaska Region Report No. 181. 
KARI, PRISCILLA RUSSELL. 1987. Tenaina 
Plantlore. Denai'ina K'et'una. An Ethno- 
botany of the Dena'ina Indians of South- 
central Alaska. 2nd rev. ed. National 
Park Service, Alaska Region, Anchorage. 



156 



GOTTESFELD 



Vol. 12, No. 1 



LITERATURE CITED (continued) 



LEIGHTON, ANNE L. 1985. Wild plant use 
by the Woods Cree (Nitihawak) of East- 
Central Saskatchewan. The National 



Museum of Man Mercury Series, Cana- 
dian Ethnology Service Paper No. 101. 



BOOK REVIEW 



Economic and Medicinal Plant Research: Volume 4: Plants and Traditional 
Medicine. H. Wagner and Norman R. Farnsworth (general editors); Olayi- 
wola Akeerele and Charlotte Gyllenhaal (volume editors) . London: Academic 
Press, 1990. Pp. xv, 174. $69.95 (hardcover). ISBN 0-12-730065-1. 



Some books are desiened as encyclopedic com 



meant 



information to spark the readers' interest and prompt 



This 



many weighty tomes would be necessary to cover every 



mer 



problem 



This book was meant to aid in the internationalization of research on medicinal 
plants, to create a dialog among countries, and to further interest on the evalua- 
tion, utilization, and conservation of medicinal plants. The body of the book 
consists of eight chapters, each dealing with a specific country (China, Ghana, 
India, Japan, Mexico, Panama, Samoa, and Thailand). The authors of the chapters 
are experts on the ethnopharmacology of that particular nation. Each author was 
asked to give a brief summary of the traditional medical system prevalent in the 
country, then to identify five or six species which seemed to merit further atten- 
tion. For each plant chosen, some important aspects of the plant's uses and con- 
stituents are discussed. The treatments of the traditional medical systems give 
general overviews of the theoretical framework in which the medicines are used 
and identifies areas for further investigation. 

The final chapter by Alexander M. Schmidt outlines the procedures used by 
the United States Food and Drug Administration in assessing the effectiveness 
and safety of new pharmaceuticals and addresses the problem of why traditional 
medicines rarely gain approval. While not following the structure of the rest of 
the book, this chapter does provide some useful background information on what 
sets the U.S. apart from other countries where herbal medicines are used more 
frequently. It also helps clarify some of the obstacles preventing the more 
widespread utilization of such plants or their extracts. 

This book cannot satisfy the need for a standard reference volume since it 
only whets the readers' appetite. This function, however, it does quite well. 



Joseph E. Laferriere 
Marion Ownbey Herbarium 
Washington State University 
Pullman. WA 99164-4309 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 157 



NEWS and COMMENTS 



BOOK REVIEWERS NEEDED.-The following titles have been received for 
review in the Journal of Ethnobiology . 

ARZIGIAN, CONSTANCE M., ROBERT F. BOSZHARDT, JAMES L. THELER, 
ROLAND L. RODELL, and MICHAEL J. SCOTT. 1989. Human adaptation 
in the upper Mississippi Valley: A study of the Pammel Creek Oneota site 
(47LC61), La Crosse, Wisconsin. The Wisconsin Archeologist Vol. 70, Nos. 
1-2. Monograph of the University of Wisconsin-La Crosse, Mississippi Valley 
Archaeology Center, Wisconsin Archeological Society, Office of Publications, 
P.O. Box 1292, Milwaukee, Wisconsin 53201. 280 pp. 

CARTER, JOSEPH G., and WILLIAM J. CARTER. 1990. Folk Dentistry: Cultural 
Evolution of Folk Remedies for Toothache. The University of North Carolina, 
Chapel Hill. 104 pp. 

1990. Herbal Dentistrv: Herbal Dental Remedies from Ancient 



Times to the Present Day. The University of North Carolina, Chapel Hill. 
77 pp. 

COHEN, MARK NATHAN. 1989. Health and the Rise of Civilization. Yale 
University Press, New Haven and London. 285 pp. 

HANSEN, JULIE M. 1991. The Palaeoethnobotany of Franchthi Cave. Excava- 
tions at Franchthi Cave, Greece, Fascicle 7. T.W. Jacobsen (general editor). 
Indiana University Press, Bloomington and Indianapolis. 280 pp. 

REVEL, NICOLE. 1990. 1991, 1992. Fleurs de Paroles. Histoire Naturelle Palawan, 



urn 



Autobiography 



An account of Her Last Years and He: 
Press, Menlo Park, California. 98 pp. 



If you would like to review any of these books and would be able to have 
your review completed within four months after receiving the book, please 



write to: 



Nancy J 



Journal of Ethnobiology 



Environmental Studies Progr 

P.O. Box 3045 

University of Victoria 

Victoria, British Columbia 



V8W 



CALL FOR MEMBERS 



We are pleased to announce the formation of the Society for Phytolith 

earch, whose goals are to disseminate new information in phytolith research 

to promote interdisciplinary contacts among researchers. Dues are $10 per 

r for regular members and $7.50 for students; members will receive The 



158 



NEWS Vol. 12, No. 1 



Phytolitharien Newsletter. Please make checks payable (in U.S. funds) to the Society 
for Phytolith Research and send to: 

John G. Jones 
Society for Phytolith Research Treasurer 



Smithsonian 



APO AA 
Miami, FL 34002-0948 



NEW PUBLICATION 



The publication Index Seminum 1991: Jardin Botdnico Estatal, Mexico is now 
available from the Universidad Autonoma del Estado de Morelos. To obtain more 



information, write to: 



Rte. M.C. Rafael Monroy M. 
Jardin Botanico Estatal 



Universidad Autonoma del Estado de Morelos 

Av. Universidad No. 1001. Chamilpa 

Cuernavaca, Morelos 62210 

MEXICO 



SOCIETY FOR ARCHAEOLOGICAL SCIENCE SYMPOSIUM: A SUMMARY 

A Society for Archaeological Science symposium— "Phytolith analysis in the 
1990s: Applications in archaeological interpretation"— was presented at the 1992 
Society for American Archaeology meeting in Pittsburgh by co-chairs Susan 
Mulholland and Amy Ollendorf (both from the University of Minnesota). The 
nine papers dealt with various aspects of phytolith analysis as applied to archae- 
ological problems, and several concentrated on the systematic information gleaned 
from cultivated plants. For example, in "Identifying rice (Oryza sativa), Poaceae, 
through phytolith analysis, ' ' Deborah Pearsall (University of Missouri-Columbia), 
Elizabeth Dinan, and Marcelle Umlauf reported on the results of an on-going 
study of wild and domesticated Oryza and related genera in Asia. And in "Phyto- 
liths in the reproductive structures of teosinte and maize: Implications for the 
study of maize evolution," Dolores Piperno (Smithsonian Tropical Research 
Institute), compared phytolith production in maize with that in teosinte. Finally, 
Marsha Baenziger and Zhao, Zhijun (both from the University of 
Missouri- Columbia) presented "Clues in the search for the millets of the past: 
Opal phytoliths and how they may tell the story," which summarized the dif- 



millet 



Interesting research also was reported on extraction proced 



sediment 



including 



the comparison of three heavy liquids— in his paper entitled "A new procedure 
for extracting phytoliths from soil." Likewise, William Middleton (University of 
Wisconsin) reported on the extraction of phytoliths from tooth calculus in his 
presentation titled "Extraction of phytoliths from prehistoric and contemporary 
camelid dental calculus." 



Summer 1992 JOURNAL OF ETHNOBIOLOGY 159 



The remaining papers of the SAS symposium highlighted some of the wide 
range of phytolith-analysis applications that are available to the archaeologist. 
In "Phytolith analysis at archaeological sites for recovery of subsistence data and 
identification of stains/' Linda Scott Cummings (Paleoresearch Laboratories) 
covered a wide range of studies, including analysis of stains, grinding stones, agri- 
cultural fields, and knives. Cesar Veintimilla (University of Missouri- Columbia) 
and Deborah Pearsall's paper, "A preliminary analysis of past vegetation in the 
Jama River valley, Manabi Province, Ecuador/' presented the results of their 
reconstruction of the vegetation in the valley, including a discussion of slash-and- 
burn methods of land clearing. In "Paleoenvironmental implications from an 
Archaic site in southwestern Chiapas: The phytolith evidence/' Cynthia Pope 
(University of Texas) and John Jones (Smithsonian Tropical Research Institute) 
presented the results from a similar study, but they focused on the results that 
are available from quick (four-hour) processing methods. Finally, Irwin Rovner 
of North Carolina State University spoke on "Phytolith Problems in Trans- 
danubian Archaeology," and he illustrated how to resolves the Festucoid (Pooid) 
phytolith shapes question with computer-assisted microscopy. 

For more information on phytolith analysis, including the Society for Phyto- 
lith Research, write to Susan Mulholland at the following address: 

Archaeometry Lab 

University of Minnesota 

10 University Drive 



MN 



BITNET 



SMULHOLL ® UMDUL / internet smulholl @ ub.d.umn 



160 



Vol. 12, No. 1 



NOTICE TO AUTHORS 

The Journal of Ethnobiology has published "Guidelines for Authors" in 
Volume 10, Number 2 (Winter 1990). Many authors will be able to prepare their 
manuscripts by consulting recent issues of the Journal . If you need a copy of the 



consult the issue of the Journal 



write 



must submit two copies of their manuscript plus the original copy 
figures. Papers not submitted in the correct format will be returned 
r. Submit manuscripts written in the English language to: 

DEBORAH M. PEARSALL, Editor 

Journal of Ethnobiology 
American Archaeology Division 

103 Swallow Hall 

University of Missouri 

Columbia, Missouri 65211 USA 

FAX: 314-882-9410 



Submit manuscripts written in the Spanish language to: 

SR. ALEJANDRO DE A VILA B, Associate Editor 

Journal of Ethnobiology 
Centro de Graduados e Investigation 

A.P. 1378 
68000 Oaxaca, Oaxaca, Mexico 



NEWS AND COMMENTS 

Individuals with information for the "News and Comments" section of 
the Journal should submit all appropriate material to Gary J. Martin, 94 Blvd. 
Flandrin, 75116, Paris, France. FAX: 33/1/45533001. 

BOOK REVIEWS 

We welcome suggestions on books to review or actual reviews from the reader- 
ship of the Journal. Please send suggestions, comments, or reviews to one of the 
Journal's book review editors. Please see inside front cover for 
addresses. 

SUBSCRIPTIONS 



names 



the Journal of Ethnobiology 



S. Fowler, Department of Anthropology, University of Nevada, Reno, NV 89557. 
Subscription rates are $60.00, institutional; $25.00 individual subscribers from 
Latin America; $25.00 students subscribers; $35.00 regular individual subscribers 
except for Latin America; Joint member (spouse; one copy of journal), add $10.00; 
Postage: $8.00 (outside of U.S.A., Canada, and Mexico). Write checks payable 
to Journal of Ethnobiology. Defective copies or copies lost in shipment will be 
replaced if written request is received within one year of issue. 



CONTENTS 



EDITOR'S VIEW 



1 



EDIBLE WOOD FERN ROOTSTOCKS OF WESTERN NORTH 
AMERICA: SOLVING AN ETHNOBOTANICAL PUZZLE 

Nancy J. Turner, Leslie M. Johnson Gottesfeld, 

Harriet V. Kuhnlein, Adolf Ceska 



1 






INFLUENCES OF MID-HOLOCENE ALTITHERMAL CLIMATES 
ON MAMMALIAN FAUNAS AND HUMAN SUBSISTENCE 
IN EASTERN WASHINGTON 

R. Lee Lyman 



37 



AN OPTIMAL FORAGING ANALYSIS OF PREHISTORIC 
SHELLFISH COLLECTING ON SAN CLEMENTE ISLAND 
CALIFORNIA 

L. Mark Raab 



63 



TWO PREHISTORIC PUEBLOAN AVIFAUNAS FROM THE 
PECOS VALLEY, SOUTHEASTERN NEW MEXICO 

Steven D. Emslie, John D. Speth, Regge N. Wiseman 



83 



FOOD TABOOS AT BUZIOS ISLAND (BRAZIL): THEIR 

SIGNIFICANCE AND RELATION TO FOLK MEDICINE 

Alpina Begossi 



117 



SHORT COMMUNICATIONS 

Eduardo O. Kohn 141 

Leslie M. Johnson Gottesfeld /.................................. 153 



NEWS 



157 



BOOK REVIEWS 34 62 , 81, 139, 156 









I" 



_£5 







VOLUME 12, NUMBER 2 



WINTER 1992 



Journal and Society Organization 

EDITOR: Deborah M. Pearsall, American Archaeology Division, 103 Swallow Hall, Uni- 
versity of Missouri, Columbia, MO 65211. 

ASSOCIATE EDITOR (Spanish): Alejandro de Avila B., Centro de Graduados e Investi- 
gacion, Instituto Tecnologico de Oaxaca, A.P. 1378, Oaxaca, Oaxaca 68000, Mexico. 
NEWS & COMMENTS EDITOR: Gary J. Martin, 94, Blvd. Flandrin, 75116, Paris, 



FAX: 33/1/44919882 
REVIEW EDITOR 



Cruz 



Bulhoes-Manguinhos, 21.041 Rio de Janiero-RJ-BRASIL. 

BOOK REVIEW EDITOR: Nancy J. Turner, Environmental Studies Program, University 
of Victoria, Victoria, B.C. V8X 3P4 CANADA. 

PRESIDENT: Paul Minnis, Department of Anthropology, University of Oklahoma, 
Norman, Oklahoma 73019. 

PRESIDENT-ELECT: Cecil H. Brown, Department of Anthropology, Northern Illinois 
University, DeKalb, Illinois 60115. 

SECRETARY/TREASURER: Catherine S. Fowler, Department of Anthropology, University 
of Nevada, Reno, Nevada 89557. 

CONFERENCE COORDINATOR: Jan Timbrook, Department of Anthropology, Santa 
Barbara Museum of Natural Histnrv ?^^Q p„« C f a n i c^i p~ ar i c-, r , fa R ar u ar =i 



California 93105. 



BOARD OF TRUSTEES 



ROBERT A. BYE, JR., Universidad Nacional Autonoma de Mexico, MEXICO; ethno- 
botany, ethnoecology . 

EDELMIRA LINARES, Universidad Nacional Autonoma de Mexico, MEXICO; ethno- 
botany. 

ELIZABETH J. REITZ, University of Georgia, USA; zooarchaeology . 

Ex officio: Past Presidents Steven A. Weber, Amadeo M. Rea and Elizabeth S. Wing; 

Permanent board member Steven D. Emslie; The Editor, President, President Elect, 

Secretary/Treasurer, and Conference Coordinator. 



ADAMS 



EDITORIAL BOARD 



JANIS B . ALCORN, Agency for International Development, USA; ethnobotany, traditiona 
agriculture, ethnomedicine . 

BRENT BERLIN, University of California, Berkeley, USA; ethnobiological classification 
medical ethnobotany. 

CECIL H. BROWN, Northern Illinois University, DeKalb, USA; folk biological classification 

DAVID R. HARRIS, University College, London, ENGLAND; ethnoecology, subsistena 
systems, archaeobotany . 

TIMOTHY JOHNS, McGill University, CANADA; chemical ecology, ethnobotany. 

HARRIET V. KUHNLEIN, McGill University, CANADA; ethnonutrition, human nutrition 

GARY J. MARTIN, Grupo de Apoyo al Desarrollo Etnico, Oaxaca, MEXICO; ethno- 
biological classification. 

DARRELL A. POSEY, Oxford University, ENGLAND; natural resource management, eth- 
noecology, ethnoentomology , tropical cultural ecology. 

AMADEO M. REA, San Diego Natural History Museum, USA; cultural ecology, zooarcha- 
eology, ethnotaxonomies. 

ELIZABETH J. REITZ, UniversitV Of Georgia IT^A- mnarrhnpnlnmi 



ILLARD 



Mexico 



Coimbra and Nancy J. Turner 



lri !c" hn fu i01 ^ * P ublished semi-annually. Manuscripts for publication, information for the "News and 
comments and book review sections should be sent to the appropriate editor on the inside back cover of this issue. 



©Society of Ethnobiology 
ISSN 0278-0771 









Journal of 

Ethnobiology 



MISSOURI BWMIICfil; 



JUL 1 4 1993 



6ARDEN UBRAR* 



VOLUME 12, NUMBER 2 



WINTER 1992 



CONTENTS 



EDITOR'S VIEW 



1 



ETHNOECOLOGY, BIODIVERSITY, AND 
MODERNIZATION IN ANDEAN POTATO AGRICULTURE 

Stephen B. Brush 



161 



THE USE OF SOUND RECORDINGS AS 
VOUCHER SPECIMENS AND STIMULUS 
MATERIALS IN ETHNOZOOLOGICAL RESEARCH 

Eugene Hunn 



187 



PREHISTORIC MEDICINAL PLANT USAGE: 
A CASE STUDY FROM COPROLITES 

Kristin D. Sobolik and Deborah J. Gerick 



203 



RECENT DOCTORAL DISSERTATIONS OF INTEREST 
TO ETHNOBIOLOGISTS: FALL 1991-FALL 1992 

Terence E. Hays and Joseph E. LaFerriere 



213 



SHORT COMMUNICATION 
Gary J. Martin and Sergio Madrid 



227 



ABSTRACTS OF PRESENTATIONS 235 



NEWS and COMMENTS 273 



BOOK REVIEWS 185 ' 198 ' 202 ' 211 ' 224 ' 225 ' 232 ' 234 ' 

271, 275, 276, 278, 279 




I teach a class called "Plants and People;" as I explain to the students the 

- mm . i - • * • 1 . 



first 



humans 
think 



we read about real people and real plants; the students fan out across Columbia 



thnobotanical 



emerge 



In the process of choosing books for this class, I've had the pleasant task of 
reviewing dozens of ethnobotanies, old and new, from Zuni Breadstuff to Nch 'i 
Wana the Big River. Many of the ones still in print now line my bookshelv< 



ethnogr 



from this anthropologist's point 



much 



them into meals and medicines 



people's relationship to the plant world changes through the seasons of the year 
and under the pressures of the changing world. 

Have you found the perfect ethnography for teaching ethnobotany or 
ethnobiology? What about films or videos which take students into the realm of 
curing and the food quest? Send your suggestions, your class syllabae, and your 
thoughts on introducing students to our field to me by September 1993. We'll 
run a feature on this topic with the list of academic programs in ethnobiology 
requested by News and Comments editor Gary Martin. Send a description of 
your program to Gary or I by the same deadline. 

Finally, I'd like to thank Chris Pulliam for his service to the Journal during 
my first 18 months as editor. Chris served as interim News and Comments editor 
and my editorial assistant; his help with bibliographic editing ^was much — - 
predated. Everyone here at the University of * " r1 '" *" "" "*" 



Missouri wishes him 



new career. 



DMP 



i 



/. Ethnobiol. 12(2): 161-185 



Winter 1992 



ETHNOECOLOGY, BIODIVERSITY, AND MODERNIZATION 

IN ANDEAN POTATO AGRICULTURE 



STEPHEN B. BRUSH 

Department of Applied Behavioral Sciences 

University of California 
Davis, CA 95616 



ABSTRACT.— Genetic erosion of traditional crops (landraces) has been predicted 
as a consequence of technological modernization and social change in centers of 
crop domestication and diversity. This paper discusses the impact of these changes 
on traditional potato diversity in the Andes of Peru. Comparative research 
between two valleys suggests that Andean farmers have maintained potato 
landraces, even as they adopt modern agricultural inputs. The emphasis on diver- 
sity in the ethnoecology of potatoes is explored as an explanation of this pattern. 
Consumption factors are particularly important in farmer conservation of biological 
diversity in potatoes. 



RESUMEN.— La erosion genetica de cultivos tradicionales (variedades criollas) 
had sido predicha como una consecuencia de la modernizacion tecnologica y el 
cambio social en los centros de domesticacion y diversidad de plantas cultivadas. 
Este trabajo discute el impacto de estos cambios sobre la diversidad tradicional 
de papas en los Andes del Peru. La investigation comparativa entre dos valles 
sugiere que los campesinos andinos han mantenido las variedades criollas de la 
papa, aun al adoptar insumos agncolas modernos. Se explora el enfasis sobre 
la diversidad en la etnoecologia de las papas como una explication de este patron. 
Los factores relacionados con el consumo son particularmente importantes en la 
conservation de la diversidad biologica de las papas por parte de los campesinos. 



RESUME.— L'erosion genetique de cultures traditionnelles (especes locales) a ete 
considdree consequence de la modernisation technologique et des changements 
sociaux dans les centres de domestication et de diversite de cultures. Cette note 
examine l'effet de ces changements sur la diversite des especes locales de pom- 
mes de terre dans les Andes du Perou. Une etude comparative entre deux valles 
mettent en evidence que les paysans andins ont preserve les especes locales meme 
si ils adoptent des varietes nouvelles. Les paysans soulignent l'importance de 
conserver la diversite des especes locales. Les differents usages de la pomme de 
terre sont particulierement importants dans la conservation biologique. 



times 



become a conservation 



noted that interspecific and infraspecific variation of crop species is not evenly 
distributed, that hyperdiversity in the form of numerous locally named varieties 
or landraces occurs in certain areas, and that these areas are the most likely centers 
of crop domestication. Vavilov emphasized the importance of centers of diver- 
sity as pools of genetic resources for crop improvement. Many agricultural scien- 



162 



BRUSH Vol. 12, No. 2 



tists who are familiar with centers of crop diversity argue that the great profu- 
sion of species and genotypes is endangered by technical progress, social change, 
and environmental factors (Hawkes 1983). Yet, we have only a rudimentary 
understanding of how specific cultural traditions maintain and influence crop 
populations in centers of crop origin and evolution. Poor understanding of the 
ecology of crop diversity weakens our ability to assess the danger of genetic ero- 
sion and to plan effective conservation programs. 

This paper questions whether the loss of biodiversity under conditions of 
agricultural modernization is as likely or widespread as purported. It examines 
the case of potatoes (Solatium spp.) in the Andes of Peru, the cradle of potato 
domestication and evolution. The paper begins with a general review of the 
ethnoecology of Andean potato agriculture. It then presents data on two Andean 
valleys that have different histories of agricultural modernization. It argues that 
diversity persists in peasant agriculture even after intensification and commer- 
cialization. While ecological and utilitarian factors are often cited as paramount 
in the maintenance of diversity (e.g., Clawson 1985), this paper concludes that 
these factors can only partially explain the practice of Andean farmers to con- 
serve high levels of potato diversity. 

One obstacle to understanding crop diversity has been that variation is at the 
infraspecific level. Burtt's advice on treating infraspecific variation is stark: "the 
best thing to do with a muck-heap is to leave it undisturbed so that it quietly 
rots down. In course of time the Code of Nomenclature will no doubt accept it as 
disposable refuse" (Burtt 1970:238). Bulmer (1970) notes that a similar aversion 
seems to prevail among ethnobiologists . Nevertheless, there are several good 
reasons for grappling with the ethnobiological treatment of plant varieties, 
especially of crops. The infraspecific level is common to many ethnobotanical 
systems (Atran 1987; Berlin 1976; Dougherty 1978), and elaborate varietal classifica- 
tion is conspicuous in some folk systems. Variety identification becomes more 
significant with industrialization and implementation of intellectual property 
rights. Perhaps most importantly, the variety level is a primary unit in the manage- 
ment of agricultural ecosystems by farmers around the world. Several case studies 
reveal the merits of drawing a more explicit connection between ethnobiology 
and farmer decision making in agriculture (e.g., Johnson 1974; Richards 1985). 
These studies reflect Bulmer' s (1974) point that one of ethnobiology ' s principal 
goals is to understand how the determination of biological species relates humans 
to the biological dimension, a point that has been reiterated in the recent empha- 
sis on "indigenous knowledge systems" (Brokensha et al. 1980). 



INTRODUCTION: ETHNOECOLOGY OF ANDEAN POTATOES 



Bertonio's (1612) dictionary of the Aymara language mentions some of the 
many names applied to the vast diversity of potatoes found in the central Andes. 
Contemporary Andean farmers still use some of the same terms. While it is 
unlikely that the varieties currently grown are biologically the same as those grown 
400 years ago, there is continuity in Andean ethnobiology. Many ethnobiologists 
and geneticists have come to believe that the biolocical diversity of Andean 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 163 



farming systems is endangered by contemporary trends, particularly the diffu- 
sion of new potato varieties (Ochoa 1975). The loss of biodiversity threatens the 
continuity of Andean agriculture that has withstood the upheaval of European 
conquest and colonization. 



dominant 



mountain 



indigenous knowledge of Andean people, especially the identification and use 



animals 



biological resources that are used (Franquemont et al. 1990; Gade 1975; Tapia 



1984). The 



model 



ment 



(Murra 1975). The cultural ecology model 



tarity as one of the organizing principles of Andean society. The complemen- 
tarity principle refers to the control and use of ecologically distinct, spatially 
separated production zones by single ethnic groups. This idea was originally 
articulated bv Murra (1975) as "verticalitv." Thomas's (1973) work on energy flow 



m 



single zones, and Golte's (1980) research suggests that multiple zone use smooths 
out labor demand, thus making labor more efficient and productive than is possible 



within a single zone. 



complementary land use is an inventory 



suited to the different physical conditions of the land: soils, temperatures, 
moisture, and evapotranspiration regimes. Describing the Vilcanota Valley of Peru, 



domesticates 



crop 



is greater than for any other crop grown there. Originally domesticated from tube 
bearing members of the Solanaceae family by Andean pastoralists, the potato he 
coevolved in the Andean environment for at least 6,000 years (Pickersgill an 
Heiser 1978). Potatoes are grown throughout most of the crop zones of the Ande: 
but they predominate in the upper zones, between 3,000 m and 4,000 m abov 
sea level, and in some areas they provide up to 70% of the calories (Ferroni 1979). 

A ■ i.l y 1 * . 1 • 1 



2n=24 to pentaploid 5n = 60) are cultivated. Some 



groups (diploid 



cosm 



very 



distinct varieties have been identified by the International Potato Center, out of 



more than 13,000 Andean accessions (Huaman 1986). Over 100 



typical 



farming 



Indigenous knowledge and diversity. —Andean potato nomenclature was first des- 
cribed by LaBarre (1947) for the Aymara of Bolivia, and folk taxonomies have since 
been described for Quechua and Spanish speaking peasants of Peru (Brunei 1975; 



Zimmerer 



taxonomic 



164 



BRUSH Vol. 12, No. 2 



im 



cation: ecology (cultivated/wild/weedy; production zone), use (edible; for boil- 
ing; for freeze drying), plant and tuber phenotype, and degree of polytypy 
(number of subclasses). The similarities of potato nomenclature across languages 
and types of production systems are notable. Both terms and taxonomy found 
in contemporary nomenclature are also evident in Bertonio's (1612) Aymara die- 



distinguish potatoes by tuber phenotype, ecology, and use. 



farmers 



ble 1 presents a schematic diagram of a folk taxonomy of potatoes from 
itral and southern highlands of Peru. At the genus level, the Solarium tuber 

- 

(papa) is distinguished from other Andean tubers such as Oxalis tuberosa 
vpaeolum tuberosum. At the species level, domesticated, wild, and weedy 
ire demarcated (mikhuna papa, atoq papa, araq papa), and frost resistant, 
ltitude, bitter types (haya papa) with high glycoalkaloid content are dif- 
ated from mid-altitude types without bitter compounds (miski papa). 
2 compares the different folk species according to the four criteria men- 



tioned above. 



TABLE 1.— Schematic diagram of Andean taxonomy of potatoes (Cusco Quechua). 



Taxonomic Level Term(s) 



Genus Papa 

Species atoq araq mikhuna haya 

papa papa papa papa 



Variety qompis rutttus ruk'i wana 



Subvariety alqa yuraq 



yuraq yana 



qompis qompis wana wana 



prim 



shape (oval, spherical, flat, long), the configuration of the tuber's "eyes 
number, location, color), skin color and pattern (white to deep purpl< 



multicolored) 



stem 



color distinguish varieties. Tuber characteristics are highly subject to somatic varia- 
tion and to environmental influences. The relationship between tuber 
characteristics and plant genotype is not well understood. 

The final level of Andean potato taxonomy is the subvariety, where the only 
contrast is between tuber colors. Black (yana) and white (yuraq) subvarieties are 
frequent, and variegated skin color is labeled alga. Skin coloration varies con- 
tinuously and is transitory in some varieties, and the subvariety label is often 
understood to be unstable. In other cases, however, subvarieties are stable and 
biologically distinct. Yuraq wana and yana wana are the two stable variants of 
a single species, S. x curtilobum. 



Winter 1992 



JOURNAL OF ETHNOBIOLOGY 



165 



TABLE 2.— Characteristics of folk species of Andean potatoes (Cusco Quechua). 



Folk Species Ecology 



Use 



Phenotype 



Polytypy 



mikhuna papa broad 



adaptability; 

mid-altitudes, 

2,500-3,700 



boiling; 

soups; 

frying 



nonbitter 
tubers; 
highly 
variable 



very 
high 



hay a papa 



frost resistant; 
high altitude, 
3,700-4,100 



processing 

by freeze- 

drying; 

chuiio 



bitter tubers 



low 



araq papa 



weedy species; 
low-medium 
altitudes, 
2,500-3,200 



boiling; 
soups 



nonbitter 



low 



atoq papa 



wild species; 
all altitudes 



not used 



small tubers 



none 



taxanomic 



that constitute intermediate ranking rather than separate taxa. Table 3 presents 
a description of six intermediate ranks that are common in the central Andes. 
These intermediate ranks are labeled and usually group several varieties and sub- 



ecology 



(u 



frying from varieties with high dry matter 
ng or steaming. Farmers also contrast m 



Modem 



smooth 



larger than local varieties that are described as chalo. Chalo is used by both 
Quechua and Aymara speakers to describe mixed collections of potatoes with 
many colors and shapes. The word appears in Bertonio's (1612) Aymara dictionary 
(cchalu), suggesting that Andean farmers distinguished mixed collections long 



modern 



The rich Andean nomenclature for potatoes is prima facie evidence for great 
diversity, and diversity at the species and infraspecific levels has been well 
documented for the Andes (e.g., Hawkes and Hjerting 1989). However, very 
little is known about the actual distribution of diversity either within or between 



measurement 



distribution in Andean potato agriculture 



complexity within 



of phenotypes and genotypes at the variety level. Somatic variation, introgres- 
sion between cultivated and wild species, and hybridization within cultivated 



166 



BRUSH 



Vol. 12, No. 2 



TABLE 3.— Intermediate folk categories at variety level in Andean potato classi- 
fication (Cusco Quechua). 



Category 



Distinguishing 
Criteria 



Description 



Contrast 



wayk'u papa 



use 



unu papa 



use 



k'usi papa 



use 



miska papa 



ecology 



chaqro/chalo phenotype 



dry/mealy potatoes to unu papa 

for boiling or 

roasting 

watery potatoes for to wayk'u papa 
frying and soups 

nonbitter tubers 
for freeze-drying 
(chuno) 



to wayk'u papa 



fast growing; 
for short season 
(maway tarpuy) 

mixed colors 
and shapes 



to unnamed category 
for long season 
(ha tun tarpuy) 

to unnamed category 
for modern varieties 
(white potatoes) 



species also pose problems for measuring diversity. Geneticists who work with 
the crop have preferred to work at the ploidy or species levels rather than at the 
variety level (Hawkes and Hjerting 1989). However, recent advances in 
biochemical characterization of potatoes (Quiros et al. 1990) may help overcome 
some of the obstacles to biological assessment of diversity. These measures rely 
on isozymes, and they focus on characteristics that are far less variable or environ- 
mentally determined than plant descriptors such as tuber shape. 

While varietal naming is a centerpiece of the Andean folk classification of 



system 



ystem 



that are only partially relevant to the biological systematics of the crop. Second, 
Andean farmers are believed to overclassify diversity (Hawkes 1947), a practice 
that is exemplified by the use of several names in a single community for a single 
type of potato and by the habit of changing names for such purposes as marketing. 
There is no evidence of a sinele, master list of names that farmers know or agree 



synonymy 



names 



Q 



seem to limit them as a general tool for measuring 



farmer segregation and identification of tubers and biochemical (isozyme) pro- 
files of tubers that reflect genotype differences. The isozyme analysis is particularly 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 167 



relevant here, since one would not expect any degree of correspondence between 
a folk taxonomy largely based on one criterion (tuber characteristics) and 
biochemical identity based on characters that are invisible to Andean farmers. 
Households are the primary management unit of selection of potato varieties and 
the primary unit for maintaining diversity. Assessment of the amount of diver- 
sity kept by different households is therefor essential to an oveall under- 
standing of diversity in the agricultural system, and this assessment can rely on 
farmer identification. Thus, research on diversity can draw directly on folk 
classification, as long as the unit of analysis is the household. 



Cultural ecology of potato agriculture.— Isbell (1978) reports that households in 
Ayacucho, Peru, initially receive gifts of seed potatoes from the couple's parents, 
but afterwards they are generally on their own in selecting and maintaining 
varieties. This pattern pertains both north and south of the Ayacucho area. 
Additional varieties are acquired through trade, purchase, gifts, and wages in 
kind. Women play an especially important role in the identification and selection 
of varieties, and women are involved in every stage of potato production: seed 
selection, production, harvest, storage, processing, and cooking. The key role 
of women has been described in Ecuador (Weismantel 1988), Peru (Allen 1988), 
and Bolivia (Johnsson 1986). Men acknowledge women's superiority in plant 
knowledge and defer to them when questions arise about potato identification. 

The Andean potato crop, both within villages and across regions, includes 
a few cosmopolitan varieties that are cultivated by virtually every household and 
many varieties that are cultivated by only a few households. Cosmopolitan 
varieties include huayro in central and southern Peru, qompis in southern Peru, 
and imilla in southern Peru and Bolivia. Approximately half of the total varieties 
in a household's inventory are these common varieties, but only a small per- 
centage of a region's varieties are common (Zimmerer 1988). Common varieties 
include both native and modern ones that are kept for different purposes; the 
native ones because of their culinary and commercial value and the modern ones 
because of their yield and acceptance in the market. Improved and native com- 
mercial varieties are often grown as monocrops in single fields or blocks within 
fields, and they may account for 70-90% of the area planted in potatoes in many 
places (Mayer 1979). The prevalence of certain native and improved varieties 
means that most of the diversity can be kept in only a small portion of the farm, 
where modern and selected varieties are not grown. This pattern is facilitated 
by the fragmentation of Andean landscapes, by complementary land use by single 
households and villages, and by the practice of cultivating numerous fields in 
the same year. 

The ethnographic literature provides strong evidence that consumption is 
critically important in maintaining diversity. Virtually every study of potato selec- 
tion refers to the importance of subtle yet elaborate contrast in taste, color, and 
texture of Andean tubers (Johns 1990; Johnsson 1986). Carter and Mamani (1982) 
note that certain varieties are prized for special meals, the most favored also 
being the most delicate and least productive. Brush (1977) describes certain 
varieties that are saved for gifts. Johnsson 's (1986) study in Bolivia discusses the 
importance of potatoes and potato diversity to the cultural identity of the Aymara. 



168 



BRUSH Vol. 12, No. 2 



His emphasis reflects Carter and Mamani's (1982:98) account of the Mauca 
family's pride and prestige in possessing seed of many rare potato varieties. The 
contribution of potato diversity to Andean identity and prestige is echoed by 
Weismantel's (1988) study of Zumbagua, Ecuador. Potatoes are not a primary 
staple for most Zumbaguan families, but they retain prestige. Serving meals 
without European introductions such as barley and fava beans is a privilege of 
more affluent families. This contradicts the popular and Eurocentric notion that 
potatoes are always judged to be inferior food to cereals. 

Intuitive logic asserts that diversity also exists because it is adaptive, leading 
to more stable production in the face of great environmental heterogeneity and 
abundant pests and pathogens. Brush (1977) and Carter and Mamani (1982) report 
that farmers recognize specific agronomic characteristics in certain varieties, such 



. While diversity may endo 1 
farmers without other means 



limit 



names 



(haya papa) where more frost resistant varieties (ruki 



However 



(miski papa), where diversity is greatest, there are 



names 



to insects, disease, or poor weather. 



more diverse collection of potatoes performs b 
e one as it is moved over the heterogeneous landscape, 
performance must ultimately be traced to the performance 



We m 



temperat 



perform 



environme 



conditions, such as soils, insect predation, or disease, have rarely been reported 
or evaluated (Hawkes and Hjerting 1989). Strong resistance to disease, insects, 
and drought is rare in domesticated potatoes, although some resistance is found 
within the predominant subspecies of the Andean region (Solanum tuberosum 
subsp. andigena) (Hawkes and Hjerting 1989). While the ethnobiology of insects 
and crop diseases of the Andes has not been specifically studied, the observa- 
tions of agronomists indicate that a folk classification of six insect species that 
attack potatoes exists (Universidad Nacional San Cristobal de Huamanga 1983). 
Plant disease taxonomy is the least developed of the folk systems. Andean farmers 
gloss several diseases under the Spanish term rancha (late blight), but they do 
not recognize several major pathogens, such as nematodes and viruses. 
Knowledge of soils has been documented (McCamant 1986), but no single matrix 
seems to exist that maps potato varieties onto soil types. 

The widely practiced sectoral rotation of fields (Orlove and Godoy 1986) is 
especially problematic to an ecological interpretation of diversity. This practice 



movement 



commun 



location, then this advantage would seem to be elimi 
iuent field rotation and the practice of growing di 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 169 



collections of potatoes together, instead of placing each variety in its special niche 
and keeping it there. Andean farmers do not emphasize site-specific adaptation 
in their nomenclature or management of potatoes, and agronomic trials suggest 
that most individual potato varieties perform equally well over a broad range of 
altitudes (Zimmerer 1991b). While the contrast at the folk species level between 
nonbitter, mid-altitude types (miska papa) and bitter, high altitude, types (hay a 
papa) is salient on several axes (production zone, relative hardiness, processing, 
and consumption), the contrast between varieties is based primarily on tuber 
phenotype. Potato farmers report that the most diverse, native varieties are also 
the least resistant to disease, insects, and the effects of poor weather. 

Diversity may have a very long-term advantage that is not specifically 
recognized in Andean folk biology or immediately apparent in the short-term. 
Long-term stability of potato production may be enhanced by having a large reper- 
toire of genotypes, some of which have particular advantage as environmental 
conditions, pests, and pathogens change over time. Under different conditions, 
varieties that are now rare may become advantageous, and thus prevalent. A large 
repertoire may seem superfluous in the short run, but it allows farmers to adjust 
to new conditions, including market demand. 



Loss of diversity.— Andean agriculture has never been a static system, and 
cultivators have long been able to accommodate new technology such as Euro- 
pean crops and animals. However, the pace of change appears to have accelerated 
during this century. Market penetration, migration, population growth, political 
reform, and new technology are now ubiquitous. Integration of local communities 
into larger political and economic systems has been present since pre-Hispanic 
times, but this integration has changed both qualitatively and quantitatively in 
the twentieth century, with the spread of capitalism, through increased popula- 
tion, and the expansion of state power, roads, and mass communication. The 
Peruvian population almost tripled between 1950 and 1990, from 7.6 million to 
21.9 million (Urban and Trueblood 1990). Rural areas have experienced far less 
growth because of emigration to urban areas, but demographic pressures are felt 
everywhere, as market systems have expanded and rural hinterlands have been 
more closely incorporated into national economics. The presence of centralized 
state power has increased through means such as agrarian reform, education, 
and regional development. Andean production is now characterized as much by 
commoditization and the acquisition of new technology as by complementarity 
and community regulation. Virtually every household now uses not only some 
Old World crops and animals but also agricultural chemicals and improved An- 
dean crop varieties. 

Modern potato varieties were first released in Peru in the early 1950s, and 
they are now found in virtually every village in the highlands. These varieties 
were specifically bred to be higher yielding, better able to utilize fertilizer, and 
resistant to specific stresses such as disease or drought. Their adoption is directly 
encouraged by agricultural extension and credit policies of the government and 
indirectly by such factors as population increase or a farmer's wish to produce 
a larger surplus for the market. Two impacts of the diffusion of modern crop 



170 



BRUSH Vol. 12, No. 2 



varieties in the Andes have been reported: increased productivity (Horton 1984) 
and loss of potato diversity (Ochoa 1975). 

The concept of genetic erosion in farming systems is based on somewhat 
simplistic bioeeographv. Adopting modern crop varieties decreases the area that 



more 



to native crops should logically reduce diversity, just as the size of islands is 



(McArthur and Wilson 



with this logic is that it assumes that farmer 



remains 



nmental 



genetic erosion does not account for cultural, economic, and envin 
buffers in agricultural systems that might protect diversity. Environmental 
heterogeneity, agronomic risk limit, market factors, and cultural factors are 



limit 



remainder of this paper will examine 



tion has occurred in Andean potato agriculture in two highland valleys of Peru. 



METHODS 



The impact on traditional crop diversity of the adoption of new varieties and 
agricultural intensification should ideally be studied in a historic framework by 
following the fate of diverse native crops as these changes occur over time. By 
all accounts, modern potatoes spread rapidly throughout the highlands after 1950. 
Unfortunately, we have neither biological nor socio-economic benchmarks from 
a period before the diffusion of these varieties. The oldest systematic and pre- 
served collections of native potatoes date only to the early 1970s, and our infor- 
mation on agricultural practices before 1950 is scanty and superficial. Without 
these historic benchmarks, comparison among regions, villages, and households 
is a valuable way to estimate the impact on diversity of such factors as commer- 
cialization of agriculture or the adoption of modern potato varieties. With this 
comparison in mind, research was undertaken in two valleys in eastern Peru. 
These valleys were chosen both because of their similarities and differences. 

Study sites.— Reconnaissance in 1978 indicated that studying valleys in Peru's 
central and southern highlands would provide contrasting records of agricultural 



modernization and commercialization 



mo 



greater integration into regional and national commercial networks. Recon- 
naissance in 1978 and 1984 suggested that the Tulumayo Valley, 50 km east of 
Huancayo in central Peru, was representative of areas with records of great potato 
diversity that had undergone extensive modernization and commercialization. 
Paucartambo Valley, 50 km east of Cusco in southern Peru, was representative 
of diverse farming systems that had experienced less modernization and com- 
mercialization. The Tulumayo and Paucartambo valleys are like many others along 
the eastern Andean escarpment. They share the traditions of complementary land 
use and community control mentioned above. Rolling upland pastures descend 
into steep and narrow valleys where crops are produced on slopes that have been 
landscaped into terrace-like fields by generations of farmers. In each valley, "pea- 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 171 



communities 



some 



munities that existed independently and others that were part of haciendas before 
the agrarian reform of 1969-1970. The community provides the modern framework 
of complementary land use, but households make most of the in-field decisions, 
such as which crop and variety to plant. Production is destined both for home 



consumpt 



from a maj 



The Tulumayo and Paucartambo valleys are also alike in their emphasis on 
potato production and in the organization of potato production. Potatoes are the 
predominant crop, accounting for 54% of the cultivated land in Tulumayo and 
47% in Paucartambo. In both valleys, households cultivate numerous small plots 
across production zones that are differentiated according to altitude, crops, 
agricultural calendar, intensity of land use, and degree of community control. 



mid 



production into a lower, short cycle (maway tarpuy) and a higher, long cycle 
(hatun tarpuy); each relies on sectoral fallow and the simultaneous cultivation 



different plots bv the same household. Finally, both the Tulumay 



fame 



familiar 



agricultural 



griculture remain in a sea of more uniform, commercial agriculture 



modern 



The major differences between the two valleys nave to ao wun euuucny 
degree of commercialization of potato agriculture, and the history of adoptior 
of modern potato varieties. A generation ago, the people of the Tulumayo were 
Quechua speakers, but today Spanish is the most widely spoken language. Labo 
migration from this valley to the mining industry in central Peru has been a signifi 



century 



Mallon 1983). Tulumay 



Quechu 



Mantaro Valley. Quechu 



nant language of Paucartambo, where ethnic identity is Indian (Allen 1988). 
Paucartambo has not experienced such a singular integrating force as massive 
labor mieration to mines. Its integration into the regional economy of southern 



m and through periodic mi 



commercial farms 



duced the fundamental shift in ethnic identity experienced in the central 



griculture between the two 



by comparing three types of potatoes: im 
varieties, and mixed native varieties. Thes 
the studv in terms of classification and m< 



farmers 



Sample and survey strategies. -The object of studying the Tulumayo and Paucar- 
tambo valleys was to model the impact on traditional potato diversity of the adop- 



lm 



Tulumay 



172 



BRUSH Vol. 12, No. 2 



ystems are distributed according to altitude, and these vary 



agricultural intensity, community 



more 



nexus between peasant communities and regional market and administrative 
systems. We assumed that better access to the principal town lowered transpor- 
tation costs and made more available modern technological inputs, including 
information and new potato varieties. The complexity of potato agriculture in the 
Andes, heterogeneity within each valley, and differences between valleys were 

_ _ _ -ft A _ _ __~ _~_ <_ _ 



determining 



sampling a reasonably large number 
;gories of potato varieties: improved, 



commercial, and mixed 



sam 



November 1984. Potato producing 
iltitude and distance (travel time) t< 



administrative 



m). middle (village at 3,000-3,500 m 



)0 m). The distances from the administrative center varied by the existenc 
[uality of roads. In the Tulumayo Valley, 14 villages were surveyed. Th 
t of these to Comas, the economic and administrative center of the Tulumay 
f, was 30 minutes by car, and the most distant was three hours over a roa< 
ad opened in 1984, the year our fieldwork began. In Paucartambo, 10 village 
surveved. The closest to the town of Paucartambo was 30 minutes by cai 



most 



The 



community 



permission, and select the sam 



communities in Peru keep mem 



questionnaire was prepared to gather information from the elected officials about 
the village inhabitants, using the membership list as the base. The village officials 
were asked to estimate age of the household head, marital status, family size, 
educational background, extent of land holdings, off farm employment, and socio- 
economic status of each member of the community. This information was then 
used to select a sample of households for the main survey. The objective was 
to include households of different socio-economic status in a survey of between 
15 and 20 households in each village. A total of 154 surveys were conducted in 
the Tulumayo Valley and 204 in Paucartambo. 

The main survey instrument was applied between January and July, 1985, 
in the Tulumayo Valley and between January and September, 1986, in the Paucar- 
tambo Valley. The survey instrument included questions on household 
characteristics and farming practices and the use of the production from each 
cultivated parcel. Questions about fallow land were also asked. We gathered 
information about each of the three general types of potatoes, improved, native 
commercial, and mixed native, such as how seed was acquired and how often, 
and advantages and disadvantages of the types. Finally, an inventory of varieties 
was made by drawing a random lot of 100 tubers from the potatoes stored in the 
farmer's house. These tubers were then sorted by the farmer into varieties that 
he recognized, and three to six tubers of each variety v/ere requested for analysis 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 173 



and conservation in the collections of the International Potato Center and the 



interviewing 



we would not be able to conduct an inventory of the entire sample of households 

- - * m * v v. «-. ft •„• _!__•• - 



were 



farm 



households that were chosen randomly from the larger sample after completion 
of the survey. In the Tulumayo Valley we obtained inventories from 87 



com 



The survey took approximately one and a half hours to complete. A survey 
team of five interviewers was employed in each valley. One surveyor worked 
in both valleys. The surveyors had university training in anthropology, economics, 
and agronomy. More than half of each team had previous fieldwork and survey 
experience. Team members lived in villages for one to three months while con- 
ducting the survey. Besides the survey on potato varieties and farming practices, 
the surveyors made ethnographic notes on each of the households that they 
surveyed. In addition to the variety survey, the research team also produced a 
land use map of each valley and completed a detailed study of the economics 



(May 



princi 



treatment 



native 



commercial, and mixed native. Improved potatoes are bred for their adaptability 
to a wide range of Andean environments. They are planted in both of the pro- 
duction zones for nonbitter potatoes, although they are more common in the 
lower one (maway tarpuy). They are usually smooth and white skinned and 
somewhat higher in water content than native potatoes. Improved varieties are 
especially important for commercial production, and they represent a large 
proportion of all potatoes sold (78% in Tulumayo and 62% in Paucartambo). 
The native commercial varieties are local varieties that have been intensively 
selected and are grown in monocultures for the market. They are regarded as 
excellent eating potatoes by farmers and urban consumers alike. They have 
moderately deep eyes, colored skin, and cream colored flesh with a high per- 
centage dry matter. These are varieties that are grown by virtually every 

household. They are well known to merchants ai ' 

commercial demand has led farmers to plant them 
managed much like fields of improved varieties. 



urban consumers, and 



Mixed 






prime source of diversity in Andean potatoes. Their tubers are usually small anc 
come in many shapes and colors, inside and out, a characteristic that is capturec 
in the label chalo. These mixed native varieties have tubers with many differem 
degrees of dryness and different flavors. They are planted as random collection; 
in the high zone (hatun tarpuy). Production from mixed native collections is sold 
but they are not subject to the same selection pressures as the specifically com 
mercial varieties. These collections are planted by almost every household in th< 
two vallevs but in only a small portion of the total land of Tulumayo farms. 



RESULTS AND DISCUSSION 
n the two valleys and among th 



number 



Table 4 



174 



BRUSH Vol. 12, No. 2 



TABLE 4.— Potato distribution in the Tulumayo and Paucartambo valleys. 
Values are percentages of potato types by zone. 



Tulumayo Valley Paucartambo Valley 

(n = 154) (n = 204) 

Native Native Native Native 

Improved Commer. Mixed Improved Commer. Mixed 



Hatun Tarpuy 
(Long Season 
Zone) 23 75 89 28 10 98 



Maway Tarpuy 
(Short Season 
Zones) 77 25 11 72 90 2 



Total 100 100 100 100 100 100 



shows the distribution of different potato types according to the weight of seed 
reported by farmers in the survey. The important contrast here is between 
improved and mixed types in the two production zones. Predictably, improved 
varieties are grown primarily in the lower zones that are farmed more intensively 
throughout the central Andes for commercial purposes (e.g., Mayer and Fonseca 
1979). Native tvpes, especiallv mixed varieties, are concentrated in the high zone 



griculture is less intensive and community control more 



commer 



varieties. In the Tulumayo Valley, these are primarily grown in the high zone 
(hatun tarpuy) while in Paucartambo, they are grown mostly in the lower zone 
(maway tarpuy). In both valleys, the lower zone is dedicated to commercial 
production of the high-yielding improved and high-value native commercial 
varieties, to take advantage of high off-season prices. The difference in location 
of native commercial varieties reflects the fact that Paucartambo farmers plant 
another commercial crop, barley, in the upper zone. Native commercial types 
are less important in Paucartambo than in Tulumayo. 

Table 5 presents data on the percentage of potato area planted in the three 
types and the use of the entire potato crop for sale and consumption for each 
valley. These figures refer to the total seed planted and harvest weight of potatoes 
as estimated by farmers. Eighty-nine percent of the Tulumayo Valley's potato 
area is planted to the more commercial types, in comparison to only 39% of the 
Paucartambo Valley. This contrast suggests that potato production in the 
Tulumayo Valley is more commercially specialized. Improved types comprise the 
bulk of potatoes that are sold in each valley. In Paucartambo the consumption 
of mixed native potatoes is considerably higher, reflecting their greater area there 
than in Tulumayo. 



data on potato use (sale, consumed, saved for seed) of eacl 
potatoes. The data in Table 6 show that improved potatoe 



Winter 1992 



JOURNAL OF ETHNOBIOLOGY 



175 



TABLE 5.— Potato farming systems in 1 
Values are percentages of potato types. 



Tulumayo Valley 



Paucartambo Valley 



(n 



154) 



(n 



204) 



Native 



Native 



Native 



Native 



Improved Commer. Mixed Total Improved Commer. Mixed Total 



Area in potatoes 

Percentage of 
potatoes sold 

Percentage 
of potatoes 
consumed 



59 



78 



41 



30 



20 



37 



11 



2 



22 



100 



100 



100 



31 



62 



30 



8 



7 



8 



61 



31 



62 



100 



100 



100 



are primarily grown for sale in both valleys, but the data also reveal that rela- 
tively high percentages of the mixed native varieties are also sold. The data 
presented in Tables 4-6 suggest that variation in use is continuous rather than 
discrete. Farmers in both valleys grow different varieties and mixes of varieties 
in separate fields, but the production from all fields is used for both consump- 
tion and sale. Comparing Tables 5 and 6 indicates that no simple division can 
be made between production for sale of improved potatoes versus production 
for use of mixed native potatoes. The consumption of native potatoes is higher 
in Paucartambo, but so is their sale, since modern potatoes represent a smaller 
proportion of all potatoes produced. 

The variation of diversity, management level, and commercialization between 
fields of different varieties and combinations of varieties is continuous. Tulumayo 
and Paucartambo farmers do not create, conceive of, or manage fields according 
to a matrix of discrete types: low input or high input, commercial or subsistence. 



TABLE 6.— Potato use by type in Tulumayo and Paucartambo valleys. Values 
are percentages of potato types. 



Tulumayo Valley 



Paucartambo Valley 



(n = 154) 

Native 



Percentage sold 

Percentage 
consumed 



80 



9 



Native 



62 



26 



45 



43 



64 



20 



(n - 204) 

Native 



Improved Commer. Mixed Improved Commer. 



50 



39 



Native 
Mixed 



25 



55 



Percentage 
saved for seed 



11 



12 



12 



16 



11 



20 



Total 



100 



100 



100 



100 



100 



100 



176 



BRUSH Vol. 12, No. 2 



The continuous gradation of management, selection, and use allows ample 
opportunity for different mixes of local and outside inputs, enabling the conser- 
vation of traditional varieties and production technology throughout the system. 



Keeping diversity. —The social framework of Andean agriculture is experiencing 
fundamental changes: incorporation of local communities into larger regional 
systems (especially markets), demographic growth, development of economic and 
technological infrastructure, and political and social restructuring of Andean 
society through land reform. It has been common to assume that such changes 
will bring about the rapid decline of diversity, as "traditional" agriculture is 
replaced by a "modern" system (Hawkes 1983). 

Improved potatoes came into the Tulumayo Valley almost as soon as they 
were released in 1950, but they did not appear in Paucartambo until 1960. 
Tulumayo farmers average almost twice as many years as their Paucartambo 
counterparts in producing these varieties (Table 7), and this longer period of adop- 
tion is reflected in the higher percentage of potato area in the Tulumayo Valley 
that is planted with improved varieties. Tulumayo suggests a glimpse of the path 
of agricultural change that Paucartambo may follow, as measured by agricultural 



TABLE 7.— Technology adoption and potato diversity. 



Tulumayo Valley Paucartambo Valley 



Average number years 

using improved varieties 1 13.9 7.8 

Year of first introduction 

of improved varieties 1 1950 1960 



Percentage of farmers who 
have ever planted improved 
varieties 2 



95.8 79.0 



Cumulative number of 

improved varieties planted 

since introduction 2 16 7 

Percentage of farms 

using purchased fertilizer 2 97.9 85 g 

Percentage of farms 

using pesticides 2 95. 7 jj 2 

Average number of native 

varieties per farm 1 12.8 9 6 



1 

2 



n = 87 for Tulumayo and 85 for Paucartambo 
n = 154 for Tulumayo and 204 for Paucartambo 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 177 



intensification, commercialization, and adoption of new technology. If we accept 
the model that the diversity of traditional crops will be adversely affected by the 
increased use of modern technology— in particular, seed varieties and increased 
integration into the market— then a comparison between the two valleys may give 
us some idea about the fate of native potatoes. 

On a biogeographical basis, we may expect crop diversity to decrease as com- 
mercial and more intensive agriculture relying on modern potato varieties takes 
hold over a larger and larger portion of the two valleys. Table 5 showed that fields 
of mixed native varieties represented only 11% of the Tulumayo's potato area. 
These fields are small islands of diversity, surrounded by biologically more 
uniform fields. The size of these islands of traditional, mixed potatoes is critical 
to conserving diversity. In Paucartambo, mixed native fields comprise 61% of the 
potato area. Biogeographically, we should thus expect Tulumayo to have much 
less diversity than the southern valley. In fact, the average number of varieties 
per household in the Tulumayo Valley is 12.8, compared to 9.6 varieties per 
household in Paucartambo. Table 7 sets this comparison in the context of adop- 
tion of improved varieties, illustrating the idea that the farmers of Paucartambo 
are at an earlier stage in the adoption process of improved potato varieties. It 
also suggests that diversity of native potatoes remains, even after adoption 
becomes virtually complete, as long as some area is planted to native potatoes. 

Tulumayo's higher average number of potato varieties per household may 
result from its history of having greater diversity than Paucartambo before the 
introduction of modern varieties. However, the southern valley is regarded by 
most potato biologists to be within the region of greatest diversity in Peru 
(Hawkes 1983). Nevertheless, Tulumayo's higher average underlines the point 
that modernization has not eliminated diversity. Statistical modeling indicates 
that the loss of diversity resulting from technology adoption may be asymptotic 
after an initial period of genetic erosion (Brush et al. 1992). The loss of diversity 
is neither simply described nor linear. Social reproduction in the Andes is best 
understood as a syncretic process whereby local and exogenous elements are 
continually combined (Allen 1988). Likewise, agricultural change in the Andes 
is not a dichotomous process of the replacement of older technology, but one 
whereby indigenous and imported technologies are combined into a single mosaic. 
Thus, fields of modern potato varieties are managed within the sectoral fallow 
system, and fields of mixed native potato varieties are rotated with European 
crops, e.g., barley and fava beans. Potato production in these two Andean valleys 
is not a dual system of production for use with native technology and produc- 
tion for sale with modern technology. Virtually every potato field has elements 
of indigenous and outside technology, and production of all types of potatoes 
is used both for consumption and for sale. 

Table 8 outlines the major reasons that favor and discourage native and 
improved potatoes in relation to three factors: consumption, commercialization, 
and production. This table suggests the complexity of determining the advan- 
tages and disadvantages of producing either type. It also shows that there is no 
single axis on which to select the two types. Cultural identity, culinary quality, 
risk, yield, and commercial demand interact in the decision. Andean farmers are 



178 



BRUSH 



Vol. 12, No. 2 



accustomed to wrestling with com 



simple dichotomies 



TABLE 8. 



improv 



Factors Favoring Selection 
Native Improved 

Varieties Varieties 



Factors Discouraging Selection 



Native 
Varieties 



Improved 
Varieties 



Consumption good tasting; lower unit 



higher unit inferior taste; 



factors 



factors 



Production 
factors 



valued for 
gifts 



cost 



costs 



less suitable for 
usual cuisine; 
larger tubers 
require more 
fuel 



Commercial higher market more profitable low yield 



value; 



(good benefit/ under 



high exchange cost ratio) 



traditional 



low market 

value; 
limited local 



value 



management market 



don't need 
new seed; 



good short- 



less 



new seed 



term resistance resistance to required 



seed readily to specific risks specific risks 
available 



Native potatoes are universally acknowledged to be culinarily superior to 



modern varieties. The hrst measure 



simpl 



time 



ment among 



! after boiling or steaming. Give the problems of synonymy, agree- 

people on the identity of different tubers, and individual taste 

differences, no single native variety represents a culinary standard for other potato 
varieties. The varieties that come closest to this status are the cosmopolitan, native 
commercial ones that are found in virtually every household, such as huayro in 
the central highlands and qompis in the south. However, I have been frequently 
told by informants that rare native varieties are equal or superior to these com- 
mercial native varieties. 

Native potatoes are preferred as a class because they are drier than the 
improved varieties, which are thought of as insipid and watery (uno 



meals 



celebrations and meals for guests emphasize native potatoes. Weismantel (198i 
writes that potatoes occupy a primary place in the system of culinary signs an 
metaphors that comprise Quichua identity in Ecuador. She observes that "White 



ti 



minor com 



served 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 179 



are favored gift items and are used to strengthen social ties, and some reports 
refer to them as "gift potatoes" (Spanish: papas de regalo) (Mayer 1979). Native 
potatoes are often expected as part of wages and during reciprocal labor exchange 
when meals are served. In the extremely tight labor market of Andean agriculture, 
the offer oiwayk'u papa as partial payment is a good way to guarantee a supply 



[ times. Ihese potatoes 
meat and wool from th 
market for all varieties 
irban and commercial 



within the vallevs is small 



do not trade, barter, or sell significant amounts of potatoes with their neighbors. 

-^ ^->-«« .« .i i • tr _ t i • J— „£— * ^l^l ******* 



Farmers 



mixed 



market 



While 



agronomically inferior to the modern 



more susceptible to the major diseases and environmental 



Paucartambo 



lm 



more susceptible to some 



of the most severe threats in Andean potato production: aphids, viruses 
and especially late blight (Phythphtora infestans) . Zimmerer (1988) poirt 
modern potato varieties have completely eliminated native varieties in 
zones of the Paucartambo Valley. 

Modern potato varieties depend on regular supplies of fresh seed tul 



farmers 



many 



fields at different altitudes. Seventy-two percent of the Tulumayo farmers and 



Paucartambo farmers in the sam 



native potato seed. Sixteen percent of Tulumayo farmers and 49% ot Faucartamoo 
farmers said they never changed improved potato seed. Large farms that are more 
commercial or farmers with more capital may thus be more able to plant modern 
varieties than capital-poor farmers. On the other hand, native potatoes are 
marketed at premium prices, and a household may broaden its economic strategy 



them for market. This is evident in the lar 



comm 



While 



diversity" remains unanswered. This 



nquiry 



why do you grow 



many types of potatoes?" is a silly and nonsensical question to Andean fanners 



Informants 



diversity is natural and a given of ft. Ande*. «2^^*~£2 



most 



complex aeroecosvstems in the world, and diversity within 



corollary of the variety of the world around them 



perspective, the question of diversity may 






180 



BRUSH Vol. 12, No. 2 



examining why farmers don't eliminate diversity in favor of a single type of potato, 
improved or native. This question is less absurd to Andean farmers, but it asks 
them to speculate about something which they don't usually do. 

Farmers in the Tulumayo and Paucartambo valleys have not eliminated diver- 
sity because they don't perceive any advantage to doing it and because there are 
advantages to keeping their mixed collections of native potatoes. Improved 
varieties don't taste very good, so they are not likely candidates to replace native 
varieties, and no single native variety meets everyone's criteria for best taste or 
best for exchange, gifts, or sale. Diversity is a pleasure in its own right when sit- 
ting down before a bowl of potatoes as the primary food at a meal. It is common 
for people to eat 20 to 30 potatoes at a single meal, and it is much more interesting 
if every other potato is a different variety. Diversity is akin to a condiment, like 
hot peppers, making meals more interesting. Naming often provides for word 
games that enliven meals. Some names are clearly evoked by the tuber's 
characteristics: pink, flat, and oval (cow's tongue, wacapahallum), cylindrical with 
eyes clustered at one end (cat's nose, tnishpasingu), or a mottled, rounded oval 
(condor egg, condor runtu). Other names evoke places, perhaps where the 
variety originated (e.g., Curimarca). This nomenclature is rich in Andean wit, 
irony, and iconoclasm. We find such folk varieties as "priest's ear" (kurapal- 



ingling), as seen through the confessional screen; another is "dog's vomit or 
"dog's stomach" (alcapapanzan) in Quechua, glossed to "Peruvian flag" in 
Spanish. 

When queried about replacing mixed native varieties with improved, higher 
yielding ones or selected native types, farmers point to two things that encourage 
diversity. First, there is no need to make such a simplifying replacement in the 
diverse Andean landscape. They cultivate potatoes in several fields each year, 
and in an Andean variant of agricultural involution, they always find space and 
time for a few native potatoes. Farmers of the Tulumayo valley have reduced this 
to a very small portion of their fields, but they maintain a high amount of diver- 
sity on this small portion. Enough potatoes are produced to satisfy local needs, 
and the market is often saturated with potatoes during the main harvest. They 
complain that they lose money on the potatoes that they do sell (Mayer and Glave 
1990). Thus there is no incentive to squeeze out the small fields of native potatoes 
for food or commerce. 

Second, the collections of mixed native potatoes are perceived as a resource 
by farmers who are economically marginalized. Mixed native potatoes are 
associated with traditional Andean agriculture and culture by subsistence and 
commercial farmers, by Quechua andmestizo farmers, and by urban consumers. 
Native potatoes are grown in the most marginal of areas by marginalized farmers. 
Like other material elements of Andean culture, such as weaving, their value has 
been inverted in the Andean kaleidoscope (Isabell 1978), that at once depreciates 
and values items of traditional culture. Products of a humiliated group, native 
potatoes command premium prices in regional markets, such as Cusco and Huan- 
cayo, where they have been elevated to the status of an artisan crop. Within 
farming communities, native potatoes are also appreciated, perhaps as much for 
their cultural significance as for their superior flavor. They are favored gift items, 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 181 



and in a rural economy that is increasingly short ot labor, they are used as added 
incentives by landowners to attract workers. 

Wholesale merchants who purchase native potatoes have a narrow concept 
of diversity and prefer the one or two varieties that have won widespread appeal 
and recognition. However, the flow of diverse native potatoes to urban markets 
is sufficient to bring new native varieties to the attention of consumers and 
merchants alike. Periodic "booms" in demand for specific varieties are sufficiently 
common to be an incentive for farmers to keep diversity as a source of seed. The 
spread of the huayro variety from the central highlands to the Cusco region and 
the local appearance of olones (Franquemont et al. 1990) fit this pattern. The 



m 



farmers 



market 



CONCLUSION 



Th 



agriculture. The ethnobiology of the potato crop emphasizes diversity at the 
infraspecific or variety level. Andean categories such as production zones (maway 
tarpuy and hatun tarpuy) and types of potato (dry, miskt papa; watery, uno papa; 
boiling, wayk'u papa) all contribute to this emphasis. This case study is represen- 
tative of several others on the maintenance of traditional crops in centers of 
agricultural origins in the face of economic and technological change in agriculture 
(Boster 1985; Brush et al. 1988; Dennis 1987; Richards 1985). These studies docu- 
ment the resilience of traditional crops, like the cultures that have produced and 
nurtured them. 



We might imagine 



m 



ment of improved and uniform 



market 



impact of this encroachment is to shrink the area devoted to mixed 
and this impact is evident in the comparison between the Tulum, 
tambo valleys. The small area of mixed native potatoes with its tre 
sity may be the last remnant of a waning agricultural system 



Ultimately, the area of native potatoes might shrink 



com 



with the European conquest 500 years ago. On the other hand, the continued 



evidence 



area and diversity may 
Andean cultural elements in the technological 



polyculture that has existed since the European conquest. 
Assuming that market incorporation, demogr 



placement 



Howev 



ggests 



diversity will survive. The disappearance of traditional crops and of diversity has 



very 



182 



BRUSH 



Vol. 12, No. 2 



Mooney 1990). Like the predicted demise 



com 



he inheritors of ancient farming traditions. Maintaining c 
survival of Latin American peasantry in the face of maj 
Janvry et al. 1989). Many factors dampen the predicted ere 
:oes in the two valleys described here. While the pressures 
lern varieties, market penetration, and population increa< 
o are cultural, economic, and environmental factors thai 



impact. There is no single axis 



farming systems change. What seems to be predictable is that farmers 



agricultural legacy. 



conserving the m 



ACKNOWLEDGEMENTS 



This research was supported by funds from the National Science Foundation (BNS 
8416724), US AID grant DPE-10680G-SS-8003, and the University of California, Davis. 
Fieldwork in Peru was conducted in collaboration with Enrique Mayer and Cesar Fonseca, 
co-principal investigators with Stephen Brush in a project on land use change in the Andes. 
Z. Huaman of the International Potato Center and R. Ortega of the National University 
of San Antonio Abad in Cusco provided useful suggestions. Research assistance in Peru 
was provided by K. Zimmerer, M. Glave, M. Granados, A. Carbajal, C. Penafiel, J. Perea, 
J. Lopez, L. Concha, E. Gudiel, and T. Inca Roca. Mauricio Bellon, Suzanne Vaupel, and 
Heather Jersild provided valuable assistance and suggestions for the analysis of field data. 
Hilda Murguia and Ramiro Ortega assisted with the Quechua orthography. Mauricio Bellon, 
Daniel Mountjoy, Benjamin Orlove, Anne Fitzgerald, Laura Merrick, and Timothy Johns 
gave useful comments on an earlier draft. Any errors are the author's responsibility. 



LITERATURE CITED 



ALLEN, CATHERINE J. 1988. The Hold Life 
Has: Coca and Cultural Identity in an 
Andean Community. Smithsonian Insti- 
tution Press, Washington, D.C. 

ATRAN, SCOTT. 1987. Origin of the species 
and genus concepts: An anthropological 
perspective. Journal of the History of 
Biology 20:195-279. 

BERLIN, BRENT. 1976. The concept of rank 
in ethnobiological classification: Some 
evidence from Aguaruna folk botany. 
American Ethnologist 3:381-399. 

BERTONIO, P. LUDOVICO. 1612. Vocabu- 
lario de la Lengua Aymara. Centro de 
Estudios de la Realidad Economica y 
Social (reprinted in 1984), La Paz, Bolivia. 

BOSTER, JAMES S. 1985. Selection for per- 
ceptual distinctiveness: Evidence from 
Aguaruna cultivars. Economic Botany 
39:310-325. 



BROKENSHA, DAVID, MICHAEL WAR- 
REN, and OSWALD WERNER. 1980. 
Indigenous Knowledge Systems and 
Development. University Press of Amer- 
ica, Lanham, Maryland. 

BRUNEL, GILLES R. 1975. Variation in 
Quechua folk biology. Unpublished 
Ph.D. dissertation, Department of An- 
thropology, University of California, 
Berkeley. 

BRUSH, STEPHEN B. 1977. Mountain, Field, 
and Family: The Economy and Human 
Ecology of an Andean Valley. Univer- 
sity of Pennsylvania Press, Philadelphia. 

, MAURICIO BELLON, and 

ELLA SCHMIDT. 1988. Agricultural 
development and maize diversity in 
Mexico. Human Ecology 16:307-328. 

, HEATH J. CARNEY, and 



ZOSIMO HUAMAN. 1981. Dynamics of 



Winter 1992 



JOURNAL OF ETHNOBIOLOGY 



183 



LITERATURE CITED (continued) 



Andean potato agriculture. Economic 

Botany 35:70-85. 

, J. EDWARD TAYLOR, and 



MAURICIO BELLON. 1992. Technology 
adoption and biological diversity in 
Andean agriculture. Journal of Develop- 
ment Economics 39:365-387. 

BULMER, RALPH. 1970. Which came first, 
the chicken or the Egg-head? Pp. 1069- 
1091 in Echanges et Communications. 
Jean Pouillon and Pierre Miranda (edi- 
tors) Mouton, The Hague. 

. 1974. Folk biology in the New 

Guinea highlands. Social Science Infor- 
mation 13:9-28. 

BURTT, B.L. 1970. Infraspecific categories 



FRANQUEMONT, CHRISTINE, TIMOTHY 
PLOWMAN, EDWARD FRANQUE- 
MONT, STEVEN R. KING, CHRISTINE 
NIEZGODA, WADE DAVIS, and CAL- 
VIN R. SPERLING. 1990. The Ethno- 
botany of Chinchero: An Andread com- 
munity in southern Peru. Fieldiana, 
Botany. No. 24. Field Museum of Natural 
History, Chicago. 

GADE, DANIEL W. 1975. Plants, Man and 
the Land in the Vilcanota Valley of Peru, 
Vol. 6: Biogeographica. W. Junk B. V., 

The Hague. 
GOLTE, JURGEN. 1980. La Racionalidad 
de la Organizacion Andina. Instituto de 
Estudios Peruanos, Lima. 



in flowering plants. Biological Journal of HAWKES, J.G. 1947. The origin and mean- 



the Linnean Society 2:233-238. 
CARTER, WILLIAM E. and MAURICIO 

MAMANI P. 1982. Irpa Chico: Individuo 

y Comunidad en la Cultura Andina. 

Libreria-Editorial Joventud, La Paz, 

Bolivia. 
CLAWSON, DAVID. 1985. Harvest security 

and interspecific diversity in traditional 

tropical agriculture. Economic Botany 

39:56-67. 



ing of South American Indian potato 
names. Journal of the Linnean Society, 
Botany 53:205-250. 

. 1983. The Diversity of Crop 

Plants. Harvard University Press, Cam- 
bridge. 

and LP. HIERTING. 1989. The 



Their 



and Evolutionary Relationships. Claren- 
don Press, Oxford. 



DE CANDOLLE, ALFONSE. 1882. Origine HORTON, DOUGLAS. 1984. Social Scien- 
tists in Agricultural Research: Lessons 



des Plantes Cultivees. Germer Bailliere, 

Paris. 

DEJANVRY, ALAIN, ELISABETH SADOU- 
LET, and LINDA WILCOX YOUND. 
1989. Land and labour in Latin American 
agriculture from the 1950s to the 1980s. 
Journal of Peasant Studies 16:398-424. 

DENNIS, JOHN V. 1987. Farmer manage- 
ment of rice variety diversity in northern 
Thailand. Ph.D. dissertation, Depart- 
ment of Rural Sociology, Cornell Uni- 
versity. University Microfilms, Ann 
Arbor. 

DOUGHERTY, JANET W.D. 1978. Salience 
and relativity in classification. American 

Ethnologist 5:66-80. 

FERRONI, MARCO A. 1979. The urban bias 
of Peruvian food policy: Consequences 
and alternatives. Unpublished Ph.D. 
dissertation, Department of Economics, 
Cornell University, Ithaca. 

FOWLER, CARY and PAT MOONEY. 1990. 
Shattering: Food, Politics, and the Loss 
of Genetic Diversity. University of Ari- 
zona Press, Tucson. 



from the Mantaro Valley Project, Peru. 
International Development Research 

Centre, Ottawa. 
HUAMAN, ZOSIMO. 1986. Conservation of 
potato genetic resources at CIP (Centro 
Internacional de Papas). CIP Circular 

14:1-7. 
ISBELL, BILLIE JEAN. 1978. To Defend Our- 
selves: Ecology and Ritual in an Andean 
Village. University of Texas Press, 



Austin. 



ALLEN. 1974. Ethnoecology 



planting practices in a swidden agricul- 
tural system. American Ethnologist 

1:87-101. 
JOHNSSON, MICK. 1986. Food and culture 
among the Bolivian Aymara: Symbolic 
expressions of social relations. Uppsala 
Studies in Cultural Anthropology 7. 
Almqvist and Wiksell International, 

Stockholm. 
JOHNS, TIMOTHY. 1990. With Bitter Herbs 
They Shall Eat It: Chemical Ecology and 
the Origins of Human Diet and Medi- 



184 



BRUSH 



Vol. 12, No. 2 



LITERATURE CITED (continued) 



in the central Andes. Journal of Ethno- 
biology 6:169-204. 

PICKERSGILL, BARBARA and CHARLES B. 
HEISER. 1978. Origins and distribution 
of plants domesticated in the New World 
tropics. Pp. 208-236 in Origins of Agri- 
culture. Charles A. Reed (editor). 
Mouton, The Hague. 

QUIROS, CARLOS, STEPHEN B. BRUSH, 
DAVID S. DOUCHES, KARL S. ZIM- 
MERER, and GORDON HUESTIS. 1990. 
Biochemical and folk assessment of varia- 
bility of Andean cultivated potatoes. 
Economic Botany 44:254-266. 

RICHARDS, PAUL. 1985. Indigenous Agri- 
cultural Revolution: Ecology and Food 
Production in West Africa. Westview 
Press, Boulder, Colorado. 

TAPIA NUNEZ, MARIO and JORGE A. 
FLORES OCHOA. 1984. Pastoreo y Pas- 
tizales de los Andes del Sur del Peru. 
Instituto Nacional de Investigacion y 
Pro morion Agropecuana, Lima. 

THOMAS, R. BROOKE. 1973. Human adap- 
tation to a high Andean energy flow 
system. Pennsylvania State University, 
Department of Anthropology, Occasional 

Papers in Anthropology No. 7. 
Papas regaladas y papas regalo: Renta- UNIVERSIDAD NACIONAL SAN CRISTO- 



cine. University of Arizona Press, 
Tucson. 

LABARRE, WESTON. 1947. Potato tax- 
onomy among the Aymara Indians of 
Bolivia. Acta Americana 5:83-103. 

LONG, NORMAN and BRIAN ROBERTS. 
1984. Miners, Peasants and Entrepre- 
neurs: Regional Development in the 
Central Highlands of Peru. Cambridge 
University Press, Cambridge. 

MCARTHUR, ROBERT H. and EDWARD O. 

WILSON. 1967. The Theory of Island 

Biogeography. Princeton University 
Press, Princeton. 

MCCAMANT, KRIS ANN. 1986. The organ- 
ization of agricultural production in 
Corporaque, Peru. Unpublished Masters 
thesis, Department of Latin American 
Studies, University of California, 
Berkeley. 

-MALLON, FLORENCIA E. 1983. The De- 
fense of Community in Peru's Central 
Highlands. Princeton University Press, 
Princeton. 

MAYER, ENRIQUE. 1979. Land Use in the 

Andes. Centro Internacional de la Papa, 
Lima. 

and MANUEL GLAVE. 1990. 



bilidad, costos e inversion. Pp. 87-120 
in Peru: El Problema Agrario en Debate. 
Alberto Chirif, Nelson Manrique, and 
Benjamin Quinandria (editors). Semi- 
nario Permanente de Investigaci6n 
Agraria, Lima. 

and CESAR FONSECA. 1979. 

Sistemas Agrarios en la Cuenca del Rio 
Cariete. Departamento de Lima, Oficina 
Nacional de Evaluacion de Recursos 
Natu rales, Lima. 



MURRA 



micas y Pollticas del Mundo Andino. 
Instituto de Estudios Peruanos, Lima. 
OCHOA, CARLOS. 1975. Potato collecting 



BAL DE HUAMANGA. 1983. Diagnos- 
tico Tecnico Agropecuario de las Com- 

unidades Campesinas de Arizona, Qas- 

anqay Ayacucho, 1983. Instituto Inter- 

americano de Cooperacion para la Agri- 

cultura (IICA), Lima. 

URBAN FRANCIS AND MICHAEL TRUE- 
BLOOD. 1990. World Population by 
Country and Region, 1950-2050. U.S. 
Department of Agriculture, Washington, 
D.C. 

VAVILOV, NIKOLAI. 1926. Studies on the 
Origin of Cultivated Plants. Institute of 
Applied Botany and Plant Improvement, 
Leningrad. 



expeditions in Chile, Bolivia and Peru, WEISMANTEL, MARY J. 1988. Food, Gen- 



and the genetic erosion of indigenous 
cultivars. Pp. 167-173 in Crop Genetic 
Resources for Today and Tomorrow. 



der and Poverty in the Ecuadorean 
Andes. University of Pennsylvania Press, 
Philadelphia. 



International Biological Programme Vol. ZIMMERER, KARL. 1988. Seeds of subsis- 



2. Otto H. Frankel and John Gregory 
Hawkes (editors). Cambridge University 
Press, Cambridge. 
ORLOVE, BENJAMIN S. and RICARDO 
GODOY. 1986. Sectoral fallow systems 



tence: Agrarian structure and genetic 
erosion in the peasant society of Paucar- 
tambo, Peru (Cusco). Ph.D. dissertation, 
Department of Geography, University of 
California, Berkeley. University Micro- 



Winter 1992 



JOURNAL OF ETHNOBIOLOGY 



185 



films, Ann Arbor. 

. 1991a. Managing diversity in 

potato and maize fields of the Peruvian 
Andes. Journal of Ethnobiology 11:23-49. 



. 1991b. The regional biogeo- 

graphy of native potato cultivars in high- 
land Peru. Journal of Biogeography 
18:165-178. 



^ 



BOOK REVIEW 



Plants for People. Anna Lewington. New York: Oxford University Press, 1990. 



Pp. vii, 232. $40.00 (hardbound). ISBN 0-19-520840-4. 



With the welcome 



/< 



many ways answers this need. With 



presentation succeeds in reaching the nontechnical reader yet it offers reliable 



ecessary 



material 



economic 



may be unrelated to mankind 



world 



The book is organized into seven sections: 
1. Starting the Day (soaps and cosmetics in general). An interesting technique 



cosmetician 



American 



with 



common 



2. 



prim 



3. 



ing, and various parts of clothing. 
From First Foods to Fast Foods dif 



stomach, from 



zonian cow trees, coffee and tea, plant foods for different peoples (rain 
forest and desert dwellers), and ending with future foods. 



4. 



House and Home 

wood, thatch, rattan and sundry stems 

leum. bamboo, and even plant materials 



This 



5. Your Very Good Health-Plants that Cure Us treats many species valuable 
for a wide spectrum of ills and discusses plants from the family medicine 
chest, medicines in the hospital, and the rain forest pharmacy. 

6. Getting Around-Plants that Transport Us spans a broad consideration from 

uses tree trunks and reeds for boats and canoes, 



numerous 



construction of docks, and plants for future fuel. 
7. Recreation-Plants that Entertain Us includes paper for words and pictures 
papyrus, wood pulp, plants for inks and dyes, the camera and plants, and 

plants in musical instruments. 
This list is far from a complete enumeration of the incredible types of employ- 



ment that the Plant Kingdom 



186 



BOOK REVIEW Vol. 12, No. 2 



m 



Religion 



many 



emisacred, a very 



culture 



feature that makes this volume a special value in teaching 



from 



U H IV XX l 4— X_ M. LAV/ V»fc-T ^^^..*^.^.— -»^--— ■ - — • — — , * 

botanical to people, products, methods, and instruments of use, temperate 



many 



mankind 



There is a comprehensive index followed by author's acknowledgements and 

picture credits. 

I must congratulate the author for her ingenuity and originality and the 

Oxford University Press for such a beautiful publication. It is indeed going to 



economic 



Richard Evans Schultes 

Director Emeritus 

Botanical Museum of Harvard University 

Cambridge, Massachusetts 



/. Ethnobiol. 12(2) -.187-198 



Winter 1992 



THE USE OF SOUND RECORDINGS 
AS VOUCHER SPECIMENS AND STIMULUS MATERIALS 

IN ETHNOZOOLOGICAL RESEARCH 



EUGENE HUNN 

University of Washington 
Seattle, WA 98195 



ABSTRACT.— The importance of collecting voucher specimens in ethnobotanical 
research is well recognized. However, collecting zoological vouchers— especially 
of large vertebrates— may prove beyond the capacity of many field projects. 
I describe the potential of field tape recordings of animal vocalizations as both 
vouchers and as stimulus materials for eliciting native terms and associated cultural 
data. Sound recordings can be at least as reliable for species documentation as 
photographs, study skins, or skeletal specimens, and such recordings are easily 
copied and edited for use in naming tasks with consultants at a later time. Basic 
equipment and procedures involved in making and using such recordings are 
also described. 

RESUMEN.— La importancia de colectar especimenes comprobantes (voucher 
specimens) en la investigacion etnobotanica ha sido ampliamente reconocida. Sin 
embargo, la colecta de especimenes zoologicos-especialmente de vertebrados de 
gran tamaho— puede estar mas alia de la capacidad de muchos proyectos de 
campo. Describo el potencial de las grabaciones de campo de vocalizaciones 
animales, tanto como especimenes comprobantes como materiales de estfmulo 
para elicitar terminos indfgerias y \ os datos culturales asociados. Las grabaciones 
de sonido pueden ser por lo menos tan confiables para documentar la identidad 
de especies como las fotograffas, pieles, o esqueletos, y tales grabaciones pueden 
ser facilmente copiadas y editadas para uso en pruebas de identificacion con 
consultores tiempo despues. Se describen tambien el equipo basico y los proce- 
dimientos necesarios para hacer y usar tales grabaciones. 

RESUME.-L'importance de la collecte des echantillons de refe'rence en 
ethnobotanique est reconnue depuis longtemps. Neanmoins, la collecte des echan- 
tillons de refe'rence en ethnozoologie, surtout pout les grands vertebres est sou- 
vent trop difficile pour la plupart des projets de recherche sur le terrain J expose 
ici l'utilite potentielle des enregistrements de vocalisations d'animaux fans dans 
la nature, aussi bien en tant qu'echantillons de refe'rence que methode servant 
a stimuler des informateurs indigenes dans l'expose de leurs savoirs b.olog.ques 
populaires et des noms concernant les animaux en question. Les enregistrements 
peuvent etre aussi utiles pour la documentation et Tetude des animaux comme 
le sont les photographies, les squelettes, ou les depouilles preservees. Et les 
enregistrements sont facilement copie's et remanies pour es etudes ulteneures 
de lexicographie a l'aide d'informateurs. Le materiel et la methode employes pour 
faire ce genre cTenregistrements sont presentes. 



188 



HUNN Vol. 12, No. 2 



The critical importance of voucher specimens in ethnobiological research has 
been repeatedly emphasized (Norton and Gill 1981; Bye 1986). As Bye notes 
(1986:2), the voucher specimen is the link between two bodies of information, 
that of Western biological science and that of the ethnoscience of the native culture 
the ethnobiologist seeks to document. For example, Sahaptin-speaking Indians 
of the Columbia Plateau employ a plant they call chalu'ksh for a variety of 
purposes, nutritional, medicinal, and as a fish poison (Meilleur et al. 1990). 
This fact remains an ethnographic particularity, however, until it can be estab- 
lished that chalu'ksh means Lomatium dissectum (Apiaceae). On the basis of this 
equation it is possible to compare a segment of Sahaptin ethnoscientific know- 
ledge with a corresponding segment of Western botanical systematics, phenology, 
ecology, and pharmacology. This equation also makes possible comparisons with 
the ethnoscientific traditions of other cultures within the range of this species. 
The resulting synthesis is of greater value than the sum of its parts, the discon- 
nected bits of ethnographic detail we would otherwise have to deal with. The 
link to Western biosystematics that the voucher establishes allows us to address 
fundamental questions, such as the nature of human knowledge itself in the 
context of human adaptation. 



WHAT IS A VOUCHER? 



specimen 



"an 



organism or sample thereof 'which physically and permanently documents data 
in an archival report by: (1) verifying the identity of the organism(s) used in the 
study, and (2) by doing so, ensure[ing] that a study which otherwise could not 
be replicated can be accurately reviewed or reassessed' [Lee et al. 1982:5]." 
To accomplish this purpose voucher specimens should meet several criteria, 
namely: (1) have recognized diagnostic characters; (2) be preserved and main- 
tained in good condition, (3) be thoroughly documented, and (4) be readily acces- 
sible in a suitable repository institution (paraphrasing Bye 1986:1). 

In the instance of vascular plant vouchers, standard operating procedures 
are well known. Basic collecting equipment such as plant presses, newsprint, 
hedge clips, pocket knives, and field dryers (or formaldehyde in the humid tropics) 
are relatively simple to obtain and use and easily transportable to the field. With 
minimal practice acceptable specimens can be produced by nonspecialists at an 
efficient rate. The accurate scientific identification of vouchers and their perma- 
nent curation, of course, require close collaboration between the field ethno- 
botanist and specialists based in established herbaria. 



The 



Bulmer 



methods). The preservation of adequate vouchers of birds in the traditional form 
of the museum study skin is a difficult and demanding skill that few ethno- 
biologists will command. Furthermore, birds are highly diverse in most of the 
world's regions, so that the ethnobiological researcher must deal with dozens, 
even hundreds of species, a number that typically exceeds by a substantial margin 
the number of all other terrestrial vertebrates combined. Birds are also elusive, 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 189 



though they may be quite conspicuous. They must first be trapped, netted, or 
shot before the onerous task of preparing their skins can begin. Furthermore, 
most are also protected from casual hunting by national regulations and by 
international treaties barring their transport across national boundaries. Col- 



permits 



museum 



misidentified 



notoriously idiosyncratic. 



specimen for a large mammal 



mention 



Similar 









be encountered by ethnobotanists, of course, as anyone who has tried to collect 
vouchers of columnar cacti or a coyol palm could attest (see Anderson 1971:227-231 
for some creative suggestions for dealing with such problems). We normally make 
exceptions to the general rule that voucher specimens are required when we are 



large 



may 



my 



this is not an adequate solution in the case of birds. 



May 



the basis of the researcher sitting down with one or two Yucatec speakers and 



Mexica 



names" for 



com 



Las Aves de Yucatan (1979)— which lists Yuc 
279 species and subspecies of birds— was 
less than satisfactory. An analysis of the Yucatec-Latin correspondences cited in 
Hartig demonstrates the need for fieldwork-based "ground truthing" to avoid 
incorrect and/or misleading attributions. Hartig began with a list of 491 bird taxa 
(species and subspecies) attributed to the Yucatan Peninsula avifauna. 1 However, 
she did not distinguish common species from rare, casual, and locally distributed 
species. This led to many overgeneralizations of native terms to species unlikely 
to have been familiar to her native consultants, such as the aplomado falcon 
(Falcofemoralis), sandhill crane (Grus canadensis), white-rumped sandpiper (Cahdns 
fuscicollis) , and gray-cheeked thrush (Cathams minimus), which are rare at best 

on the Peninsula. 

Hartig (1979) reported one to three native names for each of these 279 bird 
taxa (56.8% of the total listed for the Yucatan Peninsula). However, only 74 distinct 

(see Berlin 1992 for definitions of terms) are included in this 
nomenclatural inventory. The majority of the species "named" are labeled by 



names 



terms 



For 



exam 



>le, ch'ich'il ha' (water bird), used alone or with various ad hoc modifiers, 
is reported as the "name" of 17 different bird species representing five taxonomic 
orders (i.e., Pelecaniformes, Ciconiiformes, Anseriformes, Gruiformes, and Chara- 
driiformes). An additional 21 species, equally eclectic, are lumped I as tech ha 



seemi 



the fact that aquatic birds such as these are rarely and irregularly encountered 
on the Yucatan Peninsula except at favored coastal localities, it is unlikely that 
Hartie's main consultant-a man from a village near Vatladolid-had more than 



190 



HUNN Vol. 12, No. 2 



a casual acquaintance with most of these birds. In addition to overgeneralization 
of descriptive terms and the widespread use of nonce forms, there are numerous 
misidentifications. At least 50 nomenclatural assignments are clearly in error. 
These misidentifications appear to be due to two main factors, the consultants' 
difficulty distinguishing field guide illustrations drawn at different scales and their 



urn 



information 



is most likely the reason several bright yellow wood warblers (Parulinae, 
Emberizidae) were misidentified as one or another type of oriole (Icterus, Icterinae, 
Emberizidae), though the orioles are twice the linear dimensions of the wood 
warblers. The same difficulty may account for the equation of the Caspian tern 
(Sterna caspia) with the brown pelican (Pelecanus occidentalis) , and the confusion 
of the diminutive blue-black grassquit (Volatinia jacarina) with the bronzed cowbird 
(Molothrus aeneus). The inadequacy of pictorial representations to distinguish 
obscurely plumaged birds most likely accounts for the near random assortment 
by her consultants of diurnal raptors (Accipitridae, Falconidae), owls (Strigi- 
formes), nightjars (Caprimulgidae), and tyrant flycatchers (Tyrannidae) among 
the various named Yucatec categories appropriate to species within those larger 
groupings. Without some indication of th 



simolv inadequate as stimuli 



terminological 



A SOLUTION TO THE DILEMMA 



A simple alternative to sole reliance on pictorial stimuli presents itself: the 
use of sound recordings. Such recordings may be obtained from prerecorded 
collections when available or recorded locally in the course of the research. 
The latter is preferred as it more closely links the stimulus to the specific environ- 
mental experience of one's consultants. Field guides are strictly visual, while birds, 
especially forest birds, are far more often heard than seen. The birds themselves 
recognize one another on the basis of a variety of characteristic vocalizations 
songs and calls— rather than by sight. These characteristic vocalizations can be 
used by knowledgeable observers to identify many birds quickly and reliably at 
the level of species, and in some cases may reveal sex, age, and subspecific 
identity as well (Johnson 1982). As vocalizations are important in species recog- 
nition as well for frogs and toads, many insects, and certain mammals, the 
techniques described below are not relevant solely to ethnoornithological investi- 
gations. 

I have recently experimented with the use of field sound recordings to elicit 
ethnozoological data on birds. Serendipitously I realized that these recordings 
made excellent voucher specimens. By depositing copies of my field recordings 
in a suitable archive I met two of the four criteria Bye (1986:1) cites as necessary 
for an adequate voucher specimen, i.e., that they "be preserved and maintained 
in good condition" and that they "be readily accessible in a suitable repository 
institution." Though at present relatively few institutional repositories exist, in- 
terest in establishing such repositories is growing, with Cornell University's 



Winter 1992 



JOURNAL OF ETHNOBIOLOGY 



191 



Library of Natural Sounds the leading exemplar to date 2 (see Boswall and 
Couzens 1982 for a list of extant collections). As of 1992, the Cornell collection 
includes nearly 85,000 recordings of 4,965 animal species (Gulledge 1979). 

But what of Bye's two remaining criteria? Can they be met by sound record- 
ings? The third criterion is a matter of basic research methodoloev: that vouchers 



documented' ' means 
making the recordin 



record 



equipment used. The Library of Natural Sound provides 
i documentation forms on request, as illustrated in Fig. 



! I 



DATA 
FORM 



LIBRARY OF NATURAL SOUNDS 
CORNELL UNIVERSITY 

159 SAPSUCKER WO00S R0 
ITHACA, NEW YORK 14850 




TAXOH CODE 064- 
RECORDIST'S 

REFERENCE 

NUMBER (RRN) #69- 



*"• IDENTIFICATION and DATE (COMPLETE THIS SECTION FOR EACH DATA" FORM) 

SPECIES, SOUND, or SUBJECT : 

SAME SOUND SOURCE AS RRN: 
RECORO/STfS) #47- 

TINE (24HN) #13- 



CMSS REFERENCE TO CATALOG HO 006- 



OAY 114- 



MONTH 115- 



YEAR 116- 



RELATIVE TIME » !/-_ 
MOONLIGHT 118- NONE; 



DAWN (♦)(-) 
SOME; 



HH; 



HOW IDENTIFIED 1 20- S IGHT; 

DISTANCE TO S0UN0 SOURCE 1 57- 
BACKGR0UH0 SOUNDS #22- 



S0UN0. 

n. 



NOON (♦)(-) MR; 



IGHT. 



SUNSET (♦){-) HR 

SPECIMEN COLLECTED #19- «S; NO 



CONFIDENCE IN IDENTIFICATION #21- 
RECORDIST'S: TAPE ( ; 



CUT # 



- • .. 



.GEOGRAPHIC (EXCEPT AS NOTED. SAME AS RW: 



) 



COUNTRY or ARCHIPELAGO #07- 



STATE, 0EPT. PR0V. or ISLAND #08- 
LOCALITY #09- KM N S E 



W of 



LATITUOE #10- 



L0NG1TU0E #11- 



ALTITUDE #12 



REFERENCE FOR ANIMAL NAMES: 



'BIOLOGY AND BEHAVIOR (EXCEPT AS NOTED. SAME AS RRHT 
R OF ANIMALS #23- PRENATAL; NESTLING(S); 



FLEDGLING(S); JUVENILE(S); 



SCI 

SPECIES SOUND OR 
SOCIAL C0NT#\CT 



RANGE STATUS 



IMHATURE(S); ADULT(S); UNKNOWN AGE 



#24- 

#25- 



MALE(S); FEMALE(S); 



ISOLATED; 
FJMILT; _ 
HERD; 



INFREQUENT; 



HIXED SPECIES; 



UNKNOWN SEX 
FREQUENT; 

COLONY; 



CONSTANT; 



FLOCK; 



TROOP; 



QTHEU SOCIAL UNIT: 



#26- 



NORMAL; 

INTRODUCED; 



RANGE EXTENSION; 
CAPTIVITY 



MIGRATION; 



ACCIDENTAL; 



BREEDING STATUS 12/- NOT TERRITORIAL; 



^OUNO CATEGORY 



TERRITORIAL. 
BREEDING; 



#28- 



2ERRIT0RIAL PAIRED; 

SONG; CALL; MECHANICAL; 



TERRITORIAL SOLITARY; 



NOT BREEDING 



DEVELOPMENTAL SU8S0N6; 



OTHER SUBSONG; 



OTHER: 



SPECIAL SONG TTPE #? 9 - QUIT; COUNTER SINGING; fLIGHT; JWISPER; 



DAWN; 



MIMICRY; 



OTHER: 



STIHJLUS FOR SOUND #30- NATURAL**) PLAYBACK); SQUEAK-SPISH; HUMAN IMITATION; 



PLAYBACK OWN SONG; 



RESPONSE TO #3i- 

PLAYaAC* 

BEHAVIORAL CONTEXT #32- 
Of SOUND 



PLAYBACK SAKE SPECIES; 
OTHER: 



PLAYBACK ARTIFICIAL SOUND; 

NOME; ORIENTATION; APPROACH; NO RMAL SOMG(SOUNO); 

DIFFERENT SONG(SOUNO); ATTACK 

EXPERIMENT; ADVERTISING; COURTSHIP; COPULATION; 

MATING INVITATION; LEK; Wl CONTACT; 

NEST RELIEF; 

8E GOING; 



NEST INVITATION; 



INCUBATION; 



CARE OF YOUNG; 



ANNOYANCE; ALARM; 



PARENT -VOUK CONT*C 
THREAT; 



DISTRESS; 
FLYING; 



AGGRESSION; 



MOBBING; 



FORAGING; 



CONTENTMENT; 



ARRIVAL; 



FLOCK CONTACT; 
OEPARTURE; 



SCOLDING; 

FIGHTING; 

ROOSTING; 



FLUSHED; 



OTHER: 



SOUND DELIVERY RATE #«- 
SOUND SOURCE #34- 



SP0RA01C; 
SYR MX; 



LOU; 

LARVHX; 



NOWAL; 



HIGH; 



AIR SAC; 



VISUAL DISPLAY 
WITH SOUND 



WINGS; TAIL; 



FEET; 



BILL; 



AGITATED 

BILL DRUmiNG; 



HORNS; 



OTHER: 



#35- MO; YES(DESCRIBE IN NOTES (#67) OR VERBALLY ON TAPE) 



-^jjABjUrANO tjivlwWht M (excep t as no t ed, %** as rrn 

GENERAL CLIMATE #J6- j*T; HU MID; ARID; 

#37- TROPICAL; SUBTROPICAL; 



1 



ENVIRONMENTAL ZONE 

SEASON 

GENERAL HABITAT 



#68 
138- 



TPOPICAL. 
ARCTIC; 

SPRING; 
UOOOS; 



cmic UET^ORT 

TEMPI RATE, 



BOREAL. 



MONTANE 



HMEJtj 

FOREST, 



ALPINE, OTMER 

FAU; UINTER; 

RAW ORE ST; 



NTT; 



DRY 



RIPAR1 



TAIGA, 
tINf 



CHAPARRAL. 

SUAMP; 

DESERT; DUN 



WASSIAMO; 



JlOUOfOREST, 

SAVANNAH, 

TUNONA, MWttfi, 



•EACH, 



G*W UATER; _ mi RURA1 ■ U RBAN, 

OTHER: 



HABITAT TYPES 



#J9- 



ISLAND; 

tmiHXOtK: DECIDUOUS, EVERGREEN, SECOND GROWTH. 

UNOERGROUTH; SCRUB; THIBET, GROVE; 

ROCKT; C ANYON- RAVINE; CLIFF; 

BURROW, iALT; FRESH; BRACKJSN; ._**. 

PASTURE. MEAOOM. ORCHARD. 



SAMPV 



BAT*. CAVE. 



MUOFLAT. 

HEDGEROW; 

CULTIVATED; 



SANOSPIT; __ 
EDGE; TUSSOCK; 



FALLOW; 



YARD. 

BARREN, FUtD: 



PANK-CiMRR. 
ROADSJW. 



MOT; 



CLEARING; EXOTIC. tW»i 



DOMINANT PLANT(S) 
COVER DENSITY 
STRATA IN HABITAT 



• 40- 
•41- 
#4? 



NOME; 



SUWACf . 



OPEN; 

10H; 



SPARSE J 
MEDIUM. 



MEDIUM. 

HIGH; 



THICK 



TRUNKS -LIMBS. LOU FLIGHT. 



UATER ASSOCIATION #43- 



WEATrfEP 



#44 



LAKE; 
CLEAR, 

HIND, 



CANOPY; 
HIGH FL1GNT; S0N6 PERCH 

PCNO: 



MOUNTAIN STREAM; CREF«. tNB; _ 

L*GOON; ESTWfi*. SEA-4CEM; B*» 

ClOUOS; (WEBCAST. F«*«, RAIN. 



SNOW. 



OTHER: 



TEMPERATURE (AIR) #4S 



DEGREES (C)(F) 



MATER #46- 



«r,«ES U)(T) 



I 



* * * *JKHN1CAL { EXCEPT AS lOT TEP, SANK AS M Hj_ „ 

TAPE SPEED 1*8- CHS. FORMAT 14V TftK MONO. TW STEREO, CMSETTI 



FIELD RECORDER »S0 



NOISE REDUCTION SYSTEM 
COPV RECORDER. 



MICROPHONE IS1 



BIAS: 



. TAPE 

EQUALIZATION 



FILTERING 



TESIOESCRIK IN NOTES) 



PARABOLA (D/FL) #i>6 : 



CM 



. ^Er.rrrnc Aim catauxi*: (for inxj r_£ii<RMr_Of KArtm*l snm "til • • 

tape cAntxmr tss- rmj>; m su*c»; «^T; sp"™ ffFKT - 

iw rtfvm/; pk xmm; co Mrru mm dtsc 

cut lvgth 0^ . qualtti itO; . ^«tt AKrnrvg *m. m 00U — 

WECTBOCHAH FILED »«; {RB j W. **** °* •*• ••* ***' 

EDfTEP; • CATALOfXt>i_ 



mjns m TAT*. §ts- 

I . MTcmrriMn>: 



wo 



i i. \ tfSTf < 1^^.^T^T^.^^^^^ ^.«>«^«*^»^•^♦*«» ^ »^^»^ , VL 



#67- 



FIG. l.-Example of Data Form for recordings deposited at the Cornell University 
Library of Natural Sounds. 



information elicited from 



ctnnograpnic information elicited rrom native w««»«""' z. ■ 

tapes or subsequently elicited by reference to the native names recorded in 
response to the recorded vocalizations may be summarized in notes submitted 
to the institutional repository where the voucher copies are to be housed. This 
summary may include in addition to local names information on uses and other 



192 



HUNN Vol. 12, No. 2 



aspects of local knowledge and belief about that particular organism. Published 
analyses of these data are then substantiated by references to the repository catalog 
numbers of each voucher vocalization. 

I begin each taping session by recording in my own voice the date and time, 
location, and habitat. If consultants or colleagues are with me, that is also noted. 
I record on the same tape tentative identifications of the sounds or comments 
on the appearance or behavior of the organisms whose vocalizations I am record- 
ing as they occur. This information may be essential to verify identifications after 
the fact. I wait until the particular trip is concluded to prepare a master tape log. 
(Master tapes are my original, real time recordings; from these I may subsequently 
compose tapes arranged in systematic order, or otherwise arranged for specific 
purposes.) 

The master tape log is simply a written listing of each identifiable vocali- 
zation on the master tape in the sequence in which it has been recorded, keyed 
to the tape counter— an arbitrary and variable index of elapsed time. I record in 
this log whenever possible the presumed identity of the calling or singing bird 
(or frog, cicada, or cricket, etc.) that is most prominent during that tape segment. 
I also note bird vocalizations or other noises in the background, as this may 
provide clues to habitat associations and my affect consultants' interpretations. 
If the bird was seen at the time it was recorded and its identity confirmed by 
visual cues, this should be noted. (Vocalizations of uncertain identity should 
be confirmed visually whenever possible.) 

When a tape is subsequently reviewed by a native consultant, that consul- 
tant's identifications and comments may be keyed to the specific stimulus vocali- 
zation by reference to the master tape number and position on that tape— by side 
and elapsed time as indicated by the tape counter. Table 1 illustrates this pro- 
cedure. Ethnographic notes in this instance have been limited to native names. 



TABLE 1. 



names 



Location f/b English name Yucatec ID 



2a:012-053 fore "hammer" cricket martiyo maas 

2a: 063 back cricket sp. #2 _ 

2a:068 back mottled owl 

2a: 111-131 fore plain chachalaca baach [bach] 

2a: 135-165 fore mottled owl kul-te' 

2a: 187-215 fore mottled owl kul-te' 

2a:210 back cicada sp. #1 

2a:218+ back brown jay 



pa'ap 



2a:222+ back cicada sp. #1 ch'och* lift [chooch lin] 

2a:226-233 back blue-crowned motmot 

2a:236+ back cicada sp. 

2a: 240-244 back mottled owl 



Winter 1992 



JOURNAL OF ETHNOBIOLOGY 



193 





Location 


f/b 


English name 


Yucatec ID 




2a:249 


back 


blue-crowned motmot 


toh 




2a: 255-260 


fore 


collared forest-falcon 


koos 




2a: 260 + 

2a: 264 + 


? 

• 

back 


[cicada sp. #2 ?] 
barred forest-falcon 


chipitin 

x-k'ipch'o' (in error) 




2a: 273 


back 


cicada sp. #2 


chipitin 




2a: 282 + 
2a: 285-296 


fore 
fore 


red-throated anttanager 
black-faced antthrush 


sohlin 

beech' lu'um / tsimin 
uk'aax [syn] 




2a:301 


fore 


red-throated anttanager 


sohlin 




2a: 305 
2a:313 + 


back 
back 


thicket tinamou 
black-headed trogon 


nom 

mut [mut'] 




2a: 320 


back 


spot-breasted wren 


— 




2a: 326 


fore 


violaceous trogon ? 


— ■ 



2a: 328 
2a: 332 



back 
fore 



white-fronted parrot 
tropical gnatcatcher or 



vireo sp 



? 



ts'it-kalan-ts'e' or 
x-tatak'-che' [tatak 
che'l [in error] 



2a: 345 

2a:352+ 



back 
fore 



2a:357 
2a: 368 
2a: 369 
2a: 371 
2a: 375 
2a: 376-386 
2a: 386 + 
2a: 392 + 



fore 
fore 
fore 
fore 
back 

fore 

back 

fore 



barred antshrike 
euphonia sp. or masked 
tityra ? 
red-eyed vireo ? 

trill ? 

red- eyed vireo 
white-bellied wren 
smoky-brown woodpecl 
black-cowied oriole 
violaceous trogon 
green-backed sparrow 



chinchinbakal 



x-yankotil 
? takay [in error ?] 

yuya 

mut [mut'] 
chak tsitsi [chak 
ts'its'il [in error] 



2b: 029 + 
2b: 032 



fore 

? 



long-billed gnatwren 
[black-faced antthrush ?] 



2b: 035 
2b: 046 + 
2b: 050 
2b : 062 



back 

back 

fore 
? 



2b: 068 



back 



white-bellied wren 
black-faced antthrush 
pheasant cuckoo 
[stimulus uncertain] 

melodious blackbird 



beech' lu'um [bech' 
lu 'um] 
x-yankotil 

see 2b : 032 
x-baken-chulu 

x-takay [large fly- 
catcher sp.] 
ts'iw, corrected topich' 



194 



HUNN 



Vol. 12, No. 2 



Location 



2b: 071 



2b: 073 



2b:080-086 



2b:087 
2b: 087 
2b:094-102 



2b: 120 
2b: 123 
2b: 128 



2b: 130 

2b: 123-135 
2b: 132 



2b: 137 
2b: 140 + 
2b: 152 

2b: 153 



2b: 172 
2b: 182 



f/b 



back 



fore 



back 



back 
back 
fore 



fore 

back 

back 



fore 

back 

back 



fore 
fore 
fore 
back 



back 

fore 



English name 



Yucatec ID 



[stimulus uncertain] 



hwiido [huiro] [rose- 
throated becard ?] 

mut [muf] or uulun 
k'aax [both correct] 

large dove or pigeon sp. ? tsutsuy [Leptotila dove] 



black-headed trogon 



violaceous trogon ? 
domestic dog 
Yucatan flycatcher 



or x-chuki [chuukib] 
[scaled pigeon] 



spot-breasted wren 
brown jay 

[stimulus uncertain] 



keel-billed toucan 
[stimulus uncertai 



drum- 



pek' 

x-takay, not x-k'ok' 
[x-kok] [clay-colored 
robin] 

x-yankotil 

pa' up 

like beech' lu'um [bech' 
lu'um] [black-faced ant- 
thrush] 

punch' el 

yuya [oriole spp.] 

kolon-te' 



ming 



long-billed gnatwren 
barred forest-falcon 
white-fronted parrot 
black-headed trogon 



flycatcher sp. ? 
green-backed sparrow 



NR 

x-t'ut' lx-t'uut'] 

uulum k'aax = 

[muf] 

x-takay 

chak ts'its'i 



mut 



— 



iRecorded by Eugene Hunn in the ejido of Chunhuhub, Quintana Roo, Mexico, 17 April 1991, 
in high forest (selva mediana subperrenifolia) . Yucatec Maya identifications by Sr. Felix Medina 
Tzuc of Chunhuhub. Hunn's initial Yucatec transcriptions are compared with canonical 
forms (based on Anderson 1991) following in brackets. "Location" cites master tape number 
and side and tape counter position. "Fore" and "back" (f/b) refer to sounds in the fore- 
ground or background of the tape. "NR" indicates explicit non-recognition. "-" indicates 
no explicit comment or recognition of that vocalization. See endnote 3 for scientific names. 



AN EXAMPLE 



ril 1991 1 joined my colleagues Gene and Myra 
Mayan community in Quintana Roo, Mexico 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 195 



Gene and I are both avid birders, so we took walks at dawn in the forests and 



We 



at a season 



time 



to record a sample of their 
songs and calls. Gene eventually recorded 183 bird species within the boundaries 
of the ejido of Chunhuhub (Anderson 1991). Some few we heard but never saw. 
Most were positively identified by sight at the time or were already well known 
to one or both of us. Of course, ethnobiological research progresses far more 
rapidly when the field worker knows the subject matter well. The fact that we 
were able to identify confidently the great majority of the bird sounds on our 
tapes facilitated our questioning of consultants. Nevertheless, it is possible to use 
this technique even in ignorance of the identity of the birds recorded, just as one 
can record valuable ethnographic information concerning plant specimens of 
unknown identity, so long as the vouchers are subsequently identified by experts. 
After the morning chorus had waned, we returned to Chunhuhub and 
solicited local people willing to spend an hour or two reviewing the tapes with 



We 



mi 



farmer 



environment 



vocalizations, while the farmer, Sr. Felix Medina Tsuc proved to be expert 
indeed. He confidently named 20 species of the 32 species of birds we had been 
able to identify on one tape. He classified these 20 species into 18 Yucatec folk 
generic categories. He offered three additional Yucatec names which could not 
be positively equated to Western scientific taxa. He misidentified just three species 
(and we were uncertain as to the identities of two of these vocalizations). He did 

on six vocalizations and appeared not to have noticed them.3 Sr. 



comment 



Medina 



were never able to see. On hearing the taped call, he accurately described the 
bird's appearance, behavior, and habitat preference, then pointed it out in the 
field guide. This consultant also distinguished nomenclaturally two types of 
cicadas we had inadvertently recorded in the background as ™ e "fs * s f*? e * 
of cricket-the so-called "hammer cricket" for its sharply metallic call-that I had 

suspected of being a frog. , , . j n u..* 

Not only did he accurately identify the great majority of the taped calls but 
as we reviewed the tape he took the opportunity to expound on related birds 
which we had not encountered, providing us a detailed comparative inventory 
of owls, nightjars, parrots, doves, toucans, "blackbirds/ and orioles. In many 

. ' e '. 'K . i i-.._j. u„u~,,rir»r anH rntural significance 



binomial 



plumage 
loted. In all he offered 55 folk generic animais nam 
c names in response to some 45 minutes of tape. 

CRITERION NUMBER 4: DIAGNOSTIC CHARACTERS 

We now come to the last of Bye's (1986:1) criteri a which * > P^Pf ^ 
. u,- .:„, _..u„- j_k— „Mh romrded sounds: the voucher must nave 



most problematical when dealing witn recorucu *w^«~ 

recognized diagnostic characters." In other words, experts must be able to 



196 



HUNN Vol. 12, No. 2 



Though 



mbigu 



b may be some difficulty in associating 
normally identified on the basis of m 



to museum taxonomists 
common in the Yucatan 



icus) and the Yucatan (aka tawny-collared) nightjar (Caprimulgus [salvini] badius). 
Their distinctive calls had been confounded in the published literature— each 
attributed to the other— until the error was discovered by a group of birders 
employing just the sort of sound recording equipment I used in this study (Pier- 
son 1986). They took advantage of a further useful feature of sound recordings 
to correct this long-standing error. They played back the bird's call immediately 
with the result that the calling bird came into view, seeking to drive off the 



unwelcome com 



confirm the identities of secretive forest, marsh 



Immediate 



may 



In any case, properly curated vouchers are available for reevaluation in light of 
future advances in knowledge about birds and their vocalizations. 

It is now possible to locate expert birders familiar with the avifaunas of 

;ion of the globe. These experts may not be academic scientists 
have gained their experience as a hobbyist or by working as 
ir guides on natural history excursions. Researchers should 
endeavor to contact such individuals prior to initiating their fieldwork for advice 
on song identifications in their target area and to contact local experts able to 
confirm the fieldworker's preliminary identifications. Likewise, the number and 
biogeographical coverage available on commercial recordings of bird songs is grow- 



may 



themselves 



gaps in their own field collections. (See Boswall and Couzens 1982 and Boswall 

1 HOC C • * - - 



summaries 



— — — — — -— *— — •™fc^ - ^ m m 

An additional advantage of audiotape recordings over traditional voucher 
specimens is the ease with which they may be copied and edited. This facilitates 
professional consultations, when such are required to establish voucher identifi- 
cations positively. By contrast, ethnobotanists must collect multiple-and 



non-identical 



copies of their plant vouchers in order to have copies for circu- 
lon to taxonomic specialists. Sound recordings may be replayed any number 

times with different rnnenltaritc ,,«^i« i.„_n j ,... . • ..__ 



„™« .«w nilull llt cuu.uuiuiugicai Knowledge. Tapes maybe edited for presen- 
tation in random or nonrandom orders. Such editing requires nothing more 
elaborate than a boom box" with two heads, one for playback and one for 
recording, though it is helpful if one's equipment allows collating commentary 
between edited segments. 

In sum, sound recordings, if properly documented, meet all the essential 
requirements of voucher specimens, are relatively simple and inexpensive to 
col ect and curate and provide as well a flexible research instrument for systemati- 
cally eliciting cultural data from a representative samole of local consultants. 



Winter 1992 



JOURNAL OF ETHNOBIOLOGY 



197 



A TECHNICAL NOTE ON EQUIPMENT 



For the recordings described in this paper I used a high quality portable 



Marantz PMD-221) and a Sennheiser ME 



micr 



The 



microohone was mounted 



Wickstrom 



NOTES 



1 Barbara M. de Montes (1985) has critically reviewed this list, noting a few species that should 
not have been listed and 37 additional species that should have been included. 

20rnithology Laboratory, 159 Sapsucker Woods Road, Ithaca, New York 14850, (607) 254-2473. 

3The species he identified are: thicket tinamou (Crypturellus cinnamomeus) , collared forest- 
falcon (Micrastur semitorquatus), plain chachalaca (Ortalis vetula), large dove or pigeon sp. 
(Columba speciosa or Leptotila sp.), white-fronted parrot (Amazona albifrons), pheasant cuckoo 
(Dromococcyx phasianellus) , mottled owl (Ciccaba virgata), black-headed trogon/violaceous 
trogon (Trogon melanocephalus/T. violaceus), blue-crowned motmot (Momotus momota), keel- 
billed toucan (Ramphastos sulphuratus) , lineated woodpecker (Dryocopus lineatus), black-faced ant- 
thrush (Formicarius analis), Yucatan flycatcher (Myiarchus yucatanensis) , brown jay (Cyanocorax 
morio), spot-breasted wren/white-bellied wren (Thryothorus maculipectus/Uropsilaleucogastra) 
red-throated ant-tanager (Habia fuscicauda), Euphonia sp., melodious blackbird (Dives dwes), 
black-cowled oriole (Icterus dominicensis) . Species apparently misidentified include smoky- 
brown woodpecker (Veniliornis fumigatus) [I am uncertain if the sound he was responding to 
was produced by this species or some other in the background of the tape], trop.cal gnat- 
catcher (Polioptila plumbea) or Vireo sp. [I am uncertain of the identity of the sounds to which 
he was responding], and green-backed sparrow (Arremonops chloronotus) J^chhet^ 
called by the term presumed to name the northern cardinal (Cardrnats cardinal^ • ^ 'mUally 
misidentified the calls of the barred forest-falcon (Micrastur rufrcolhs as he squ ^cuckoo 
(Piaya cayana), but later changed his mind to declare the sound unfamiliar. In fact »s pos^ble 
that the barred forest-falcon is extremely rare in this part of the Yucatan , Pemnsu ^ freoes 
for which no names were offered include a backgrounded barred ^ shr ^^cJuJ 
doliatus), long-billed gnatwen (Ramphocaenus melanurus), and red-eyed ^^atoc««>. 
Vocalizations named in Yucatec but not identified scientifically include like beech luum 
"like the black-faced antthrush"; hwiido for what may have been ca Ms of «^*™^ 
becard (Pachyramphus aglatae), and yuya for what may have been a second species of oriole 

(Icterus sp.). 



LITERATURE CITED 



ANDERSON, EDGAR. 1991. Plants, Man 
and Life. University of California Press, 



Berkeley. 



EUGENE 



hub animal names. Manuscript on file, 
Department of Anthropology, University 
of California, Riverside. 
BERLIN, BRENT. 1992. Ethnobiological Clas- 
sification: Principles of Categorization 
of Plants and Animals in Traditional 



Societies. Princeton University Press, 
Princeton, New Jersey. 
BOSWALL, JEFFREY. 1985. Bird sound 
publication in North America: An up- 
date. American Birds 39:355-356. 

and DOMINIC COUZENS. 



1982. Fifty years of bird sound publi- 
cation in North America: 1931-1981. 
American Birds 36:924-943. 
MFR RALPH N.H. 1969. Field Methods 



198 



HUNN 



Vol. 12, No. 2 



LITERATURE CITED (continued) 



in Ethno- zoology with Special Reference 
to the New Guinea Highlands. Univer- 
sity of Papua and New Guinea, Depart- 
ment of Anthropology and Sociology. 

BYE, ROBERT A., JR. 1986. Voucher speci- 
mens in ethnobiological studies and pub- 
lications. Journal of Ethnobiology 6:1-8. 

DAVIS, THOMAS H. 1981. The microphone 
comes first: Cassette tape-recording up- 
date. Birding 13:161-163. 

de MONTES, BARBARA M. 1985. Critique^of 
Las Aves de Yucatan by Helga-Maria Har- 
tig. Manuscript on file, Department of 
Anthropology, University of Washing- 
ton, Seattle. 

GULLEDGE, JAMES L. 1979. The library of 

natural sounds at the Laboratory of 

Ornithology, Cornell University. Record- 
ed Sound 74-75:38-41. 

HARTIG, HELGA-MARIA. 1979. Las Aves 

de Yucatan. Fondo Editorial de Yucatan, 



Mexico. 



SCOTT 



Beyond species recognition. American 
Birds 36:944-947. 
LEE, W.L., B.M. BELL, and J.F. SUTTON 
(editors). 1982. Guidelines for Acquisi- 



tion and Management of Biological Speci- 
mens. Association of Systematics Collec- 
tions, Lawrence, Kansas. 
MEILLEUR, BRIEN A., EUGENE S. HUNN 



RACHEL 



1990. Lomatium 



dissectum (Apiaceae): Multi-purpose plant 
of the Pacific Northwest. Journal of 
Ethnobiology 10:1-20. 

NORTON, HELEN H. and STEVEN J. GILL. 
1981. The ethnobotanical imperative: A 
consideration of obligations, implications 
and methodology. Northwest Anthro- 
pological Research Notes 15:117-134. 

PETERSON, ROGER TORY, and EDWARD 
L. CHALIF. 1973. A Field Guide to 
Mexican Birds. Houghton Mifflin Com- 
pany, Boston. 

PIERSON, JAN ERICK. 1986. Notes on the 
vocalizations of the Yucatan poorwill 
(Nyctiphrynus yucatanicus) and tawny- 
collared nightjar (Caprimulgus salvini). 
MBA "bulletin board": Newsletter of the 
Mexican Birding Association 1:3-4 (Octo- 
ber). 

WICKSTROM, DAVID. 1988. Tools of the 
trade: Bird recording equipment. Birding 
20:262-266. 



BOOK REVIEW 



The Origins of Agriculture and Settled Life. Richard S. MacNeish. Norman 
versity of Oklahoma Press, 1992. Pp. xix, 433. $75.00. ISBN 0-8061-22 



more assiduously in more aericultural horn 



Richard MacNeish. Given MacNeish's experience, and his creative and original 
mind, it is not surprising that he has produced a benchmark work on agricultural 
origins. 

This book is not a full review of the "origins of agriculture" literature, 
though MacNeish does provide a 38-page bibliography. Rather, the book 



model of agricultural development that MacNeish 



American 
MacNeish 



Mexico 



from the simplistic "Donulatinn nrpssure" mo 



of Cohen (1977) or Rindos's reduction of domestication to a virtually accidental 



(Rindos 1984). MacNeish 



many 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 199 



MacNeish allows four primary areas: Peru, Mexico, t 
time), and north-central China. This parsimonious 



Bluml 



(M 



randomly 



tions to farming. MacNeish has pointed out that, in all primary areas, domesticated 
crops produced only about 5% of the food supply for literally thousands of years. 



m because of huneer would change much 



that— or else starve 



meeafauna became 



smaller 






climates emerged. In such areas, affluent hunting groups became what MacNeish 
calls "destitute foraging bands/' It seems that these bands were not really 
destitute: thev had to shift from mobile hunting to more local seasonal-round 



macroband/microband alternation. They 



domesticate 



primary 



emerged, if there was intensive contact between groups 
and foods. 



meet all these criteria. Much 



homog 



domesticate. I believe MacNeish 
> southern South America and so 



are culs-de-sac, lacking the necessary intensive contact between groups. 

Secondary domestication took place in areas nearby, where people < 

satisfy their immediate needs without moving seasonally. They had less ii 



mov 



tion increase in the more circumscribed of these spots-notably river valleys in 
dry areas— would lead to resource overuse, and thus to borrowing agriculture. 
Readers will be reminded of Carneiro's "circumscription hypothesis" (Carneiro 
1970), as well as of Mark Cohen's work in Peru (Cohen 1977). 

Tertiary domestication is yet another process. This involved a moving fringe 
of agriculture expanding slowly over the landscape. The landscape is thus 
predicted to be relatively homogeneous. It will also have fewer easily domestic- 
able plants-i.e., it will probably be a forest or a shrub desert. Europe is the 
obvious type case, but something similar seems to have happened in Japan, 
southwestern and eastern North America, and other areas. This has previously 
been explained as natural diffusion of a good idea, or as expansion of one popula- 
tion group-a theory argued for Europe by Cavalli-Sforza and associates (Ammer- 



man 



American areas.) MacNeish 



different process; sedentary groups built up population, borrowed a few 
domesticates, traded food, built up more population, and had to intensify further. 



emerges 



grows 



are less able to pick up and move, and they are more skilled at plant breeding 



200 



BOOK REVIEW Vol. 12, No. 2 



and domestication. Intensifying agriculture becomes a more and more persuasive 
option. MacNeish tests the evidence against available world data, including some 
50 key sequences from key areas. He predicts that more sequences will only 
strengthen his case. 



Some 



domestication coincides with the arrival of warmer 



climates 



microeconomist advising a hunting band would have told them 



more mobile 



the payoff lay in getting 



more meat 



game drastically reduced and plants increasing, the advisor would recom 



me 



guinea pigs, llamas ...) and grain futures. This is what happened. We need not 
invoke population pressure— common sense, perhaps glorified as "optimal forag- 
ing theory," will do. 

It is well to remember, also, Sauer's point that the original domesticators must 
have been affluent enough to try a process that is highly chancy and slow to pay 



time 



only 



to understand. The same phenomenon, today, explains the fact that almost all 
agricultural research and innovation is in the richest nations. 

The importance of MacNeish's "intensive interactions" among the primary 
producers also deserves more attention— as MacNeish has said in other papers 
(e.g., MacNeish 1977). People can find something to domesticate almost anywhere, 
and there are many diverse habitats that require humans to move around. More 
critical is a source of ideas and some pressure to use them. Trade provides the 
ideas and some incentive. People would want to raise crops near the house to 
have them easily available for trade. This may explain why so many early cultigens 
were portable luxuries such as chile peppers, bottle gourds, and tobacco. Raid 
also provides incentive. It makes people desire to raise all their food within a 
defensible perimeter. Warfare is common in traditional societies. It keeps popula- 
tion growth rates down, and creates pressure for food concentration and storage. 
Thus it can substitute for population pressure as a source of strong incentive to 
intensify agriculture. Surely, it is no accident that the four primary centers and 
most of the secondary centers are in areas that have always been "crossroads 
of continents." 



MacNeish's model receives independent verification from 



some 



questions MacNeish asks of the data have been resolved recently, and MacNeish's 
model wins. The earliest agriculture known is in an area of dry valleys and lusher 
mountains, with many ecozones not quite accessible from one site (personal 
observation). In the latter case, McCorriston and Hole (1991) have recently 
argued for the Jordan Valley as the origin point, but most of the earliest crops 
find their closest living relatives in the area of the Turkey-Syria-Iraq tri-point 
(Giles Waines, personal communication). Either areas fits MacNeish's model for 
the Near East. 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 201 






The main problem with MacNeish's book is the high number of errors. The 
proofreading, editing, and production of this book do no credit to the University 
of Oklahoma Press. K.C. Chang's The Archaeology of Ancient China appears as The 
Archaic of Ancient China, and is cited to a long-superseded edition. Scientific 
and common names in the tables of cultivated plants are very frequently mis- 
written. 

Three controversial matters should be mentioned. First and least important, 
MacNeish separates teosinte from true "corn" (p. 110), unlike most contemporary 
scholars. Second, MacNeish credits Europe with spelt, millet, and apples— all quite 
possibly domesticated in Asia. Third, MacNeish is famous for his acceptance of 
very early dates for the entry of humans into the New World. This does not 
affect his arguments in this book, but more than a few archaeologists will be put 
off by the background sections of the chapters on New World areas. 

This book provides the best model yet published for the origins of agriculture. 
Its author's formidable experience with the data makes him worth the serious 
attention of anyone interested in the question. 



LITERATURE CITED 



AMMERMAN, ALBERT, and L. CAVALLI-SFORZA. 1984. The Neolithic Transition 
and Genetics of Populations in Europe. Princeton University Press, Princeton, NJ. 

BLUMLER, MARK A. 1992. Independent inventionism and recent genetic evidence on 
plant domestication. Economic Botany 46:98-111. 

CARNEIRO, ROBERT. 1970. A theory of the origin of the state. Science 169:733-738. 

CHANG, KWANG-CHIH. 1986. The Archaeology of Ancient China. Yale University 

Press, New Haven, CT. 
COHEN, MARK R. 1977. The Food Crisis in Prehistory. Yale University Press, New 

Haven, CT. 
HO, PING-TI. 1988. The origins of Chinese agriculture. Paper presented at the Fifth 

International Conference on the History of Science in China. 
MACNEISH, RICHARD S. 1977. The beginning of agriculture in central Peru. Pp. 753- 

810 in Origins of Agriculture. Charles Reed (editor), Mouton, The Hague 
MCCORRISTON, JOY and FRANK HOLE. 1991. The ecology of seasonal stress and 

the origins of agriculture in the Near East. American Anthropologist 93:46-69. 
RINDOS, David. 1984. The Origins of Agriculture. Academic Press, New York. 
SAUER, CARL O. 1969. Agricultural Origins and Dispersals. MIT Press, Cambridge, MA. 



E.N. Anderson 



Anthropology 



University of California 
Riverside, CA 



202 



BOOK REVIEW Vol. 12, No. 2 



BOOK REVIEW 



Whitmore 



don Press, 1990, and New York: Oxford University Press (reprinted with 



$35 



There seemed 



mankind's impact upon them 



man 



aim 



provide an introduction to the world's tropical rain forests ... to describe 
their structure and functioning, their value to man and what he is doing 
to them. 



These expectations of the author have indeed been fully achieved in this 
valuable book. Only a few such books by authorities have appeared that equal 
this volume, which is condensed but inclusive. The use of tables and charts does 
much to present an extraordinary amount of information which otherwise would 
have required pages of text. But even the text is terse, clean-cut, and always 
pertinent. This characteristic, plus the author's extensive experience, makes the 
book valuable for general readers newly interested in conservation as well as 
experts in the field; as a former university teacher, I recommend the book 
especially as a text for courses in rain forest conservation or even conservation 
of other forests in the tropics. 

The book is divided into ten chapters: (1) An introduction to tropical rain 
forests; (2) What are tropical rain forests?; (3) Plant life; (4) Rain forest animals; 
(5) Relationships between plants and animals; (6) Tropical rain forests through 
time; (7) Forest dynamics; (8) Nutrients and their cycles; (9) Species richness; and 
(10) Tropical rain forests yesterday, today, and tomorrow. There follows an 
epilogue with notes, references, a glossary, a general index, and an index of plants 
and forest products. 

As a botanical explorer of the northwest Amazon for 47 years, I find this book 
exceedingly useful and a welcome addition to my own extensive library on 



Amazonia. 



Richard Evans Schultes 

Botanical Museum of Harvard University 

Cambridge, MA 02138 



/. Ethnobiol. 12(2) :203-211 



Winter 1992 



PREHISTORIC MEDICINAL PLANT USAGE: 

A CASE STUDY FROM COPROLITES 



KRISTIN D. SOBOLIK 
artment of Anthropology 



forQ 



)f Ma ine 



Orono, ME 04469 

and 

DEBORAH J. GERICK 

Department of Anthropology 

Texas A&M University 
College Station, TX 77843 



ABSTRACT.— Identifying medicinal plant usage from the prehistoric record is 
problematic due to the preservation of such information, and the lack of con- 
clusive evidence provided through dietary and non-dietary assemblages. One 
of the most direct methods of determining prehistoric medicinal plant usage is 
through the analysis of coprolites. This paper presents data gained through an 
analysis of 32 Archaic coprolites from Caldwell Cave, Culberson County, in west 
Texas. The coprolites were analyzed for their pollen content in an attempt to 
identify flowers and/or inflorescences ingested purely for their medicinal value. 
High frequencies of Ephedra (Mormon tea) and Prosopis (mesquite) pollen were 
observed in the coprolites, and since these plants are known as diathetics, it 
is postulated that the Archaic peoples of Caldwell Cave probably used these plants 
as medicinal diarrhetics. 

RESUMEN.-La identificacion del uso de plantas con fines medicinales en el 
registro prehistdrico es problematica dada la preservacion de tal informacion y 
la carencia de evidencia concluyente proporcionada por colecciones dieteticas y 
no dieteticas. Uno de los metodos mis directos para verificar el uso prehistonco 
de plantas medicinales es el analisis de coprolitos. Este trabajo presenta informa- 
ci6n generada a travls del analisis de 32 coprolitos arcaicos de la cueva Ca dwell, 
condado de Culbertson, en el occidente de Texas. Los coprolitos fueron anahzados 
por su contenido de polen, en un intento de verificar la presencia de flores y/o 
inflorescenias ingeridas puramente por su valor medicinal. Se observaron frecuen- 
cias altas de polen de Ephedra (canutillo o tepopote) y Prosopis (mezquite). Puesto 
que estas plantas son conocidas como diarreicas, se postula que las gentes del 
periodo arcaico de la cueva Caldwell probablemente usaron estas plantas como 
diarreicas medicinales. 



204 



SOBOLIK and GERICK Vol. 12, No. 2 



RESUME.— A cause de la rarete de leur conservation et l'absence d'evidence 
conclusive a partir d'assemblages alimentaires et non-alimentaires, il est difficile 
de fournir temoignage sur l'emploi des herbes pharmaceutiques dans la 
prehistoire . L'analyse des coprolithes humains est une des methodes les plus 
directes pour determiner l'usage prehistoriques des herbes medicinales. Cette 
etude presente des resultats obtenus par l'analyse d'un echantillon de 32 copro- 
lithes humains decouverts dans la Caverne de Caldwell, un gisement prehis- 
torique datant de la periode Archaique et situe dans le Departement de Culber- 
son au Texas Occidental (Etats Unis). Le pollen trouve dans les coprolithes a e'te 
etudie de facon a determiner la presence de fleurs ou d' inflorescences consom- 
mees seulement pour leur valeur medicinale. Des pollens de genres Ephedra (the 
des Mormons) et Prosopis (mesquite) ont ete observes frequemment darts les 
coprolithes. Ces herbes sont connues actuellement comme medicaments contre 
la diarrhee. Par consequence, il est suggere que les peuples Archaiques de la 
Caverne de Caldwell ont probablement consomme ces herbes comme 
medicaments contre la diarrhee. 



INTRODUCTION 



Document 



difficult to distinguish between plants that were consumed for dietary purposes 
and those consumed for medicinal purposes only. Many times plants were 
probably used both as foods and as medicines. The analysis of plant remains from 



dietary 



remains 



archaeological sites can be deposited through a number of channels, most 
significantly through contamination from outside sources (i.e., water, wind, matrix 
shifts, and animals). Prehistoric people used plants to produce clothing, shelter, 
basketry, and twining. Plants used for these purposes become deposited into 
archaeological contexts and can be mistaken for food or medicinal items. 

One of the best methods to determine prehistoric medicinal and dietary con- 
sumption directly is through the analysis of human coprolites (dessicated human 
feces)(Bryant 1974). The analysis of coprolites can provide information on 



medicinal 



meals that were consumed 



macroremains) or up to one month before (i.e., pollen)(Kelso 1976; Williams-Dean 
1978; Sobolik 1988). 

In this paper we present information gained from the analysis of the pollen 
content of 32 coprolites recovered from Caldwell Cave, Culberson County, western 
Texas. Analysis of the coprolites reveals a direct correlation between the presence 
of plants useful for alleviating diarrhea and coprolites which were diarrhetic. This 
correlation suggests that the prehistoric population of Caldwell Cave was ingesting 
medicinal plants to help alleviate chronic diarrhea. These plants include Ephedra 
(Mormon tea) and Prosopis (mesquite). This analysis confirms the inconclusive 
evidence provided through the study conducted by Holloway (1983) which 
indicated that the Caldwell Cave occupants were possibly using Ephedra and 
Larrea (creosote bush) in medicinal teas to help cure chronic diarrhea. 



Winter 1992 



JOURNAL OF ETHNOBIOLOGY 



205 



BACKGROUND 



Caldwell Cave (41CU1) is a sink-hole located in the east central portion of 
Culberson County (Fig. 1). Caldwell Cave was occupied from approximately 
A.D. 1200-1450 (Tanner 1949). The prehistoric peoples occupying Caldwell Cave 




FIG. 1. -Location of Caldwell Cave, Culberson County, Texas. 



206 



SOBOLIK and GERICK Vol. 12, No. 2 



and the surrounding area were hunter-gatherers throughout this time period 

(Jackson 1937; Tanner 1949). 

Past and present observations of the coprolite samples indicate that chronic 
diarrhea was a prevalent health problem for the Caldwell Cave occupants. Chronic 
diarrhea may have been caused by high gypsum content in the local water 
supply— there is a gypsum stratum underlying the limestone surface of the area 
(Sayles 1941). Chemical analysis of a local spring revealed a high content of 
magnesium, which is a common ingredient in laxatives (Holloway 1983:Table 5). 

Although analyses of coprolites from Culberson County are scarce, a few 
studies have been conducted: Holloway (1983) analyzed eight coprolites from 
Caldwell Cave, and Caldwell and Murry (Murry 1980) analyzed 16 coprolites and 
the colon contents of a mummified child from Granado Cave (41CU8). Holloway 
(1983) suggested that Mormon tea and creosote bush pollen were ingested as 
medicinal diarrhetics. However, Mormon tea pollen was observed in this study 
at a low frequency (10.7%), indicating that Ephedra pollen may not have been 
intentionally ingested. The sample which contained Ephedra pollen was not diar- 
rhetic. One sample also contained Larrea pollen at a frequency of 18.7% indicating 
that this plant may have been ingested intentionally. The sample which contained 
Larrea pollen was diarrhetic. Holloway (1983) also observed high frequencies of 
grass pollen and seeds in most of the samples, high percentages of sagebrush 
pollen in two samples, and a high percentage of Hydrophyllaceae (water leaf) 
pollen in one sample. Murry (1980) also observed high frequencies of Mormon 
tea pollen (above 95%) in two of the samples he analyzed. 



METHODS AND MATERIALS 



Hamilton 



macroremain content. The sam 



m 



by Holloway (1983:Table 1): type I, solid, firm, well-formed; type II, soft, 
well-formed; type III, soft, not well-formed; and type IV, soft, paddy-like, 
apparently runny. Because it proved difficult to distinguish morphological 



combined into one category, II/III. These 



ermining 



which coprolites were diarrhetic. 

The coprolites were processed following standard techniques (Fry 1985) and 
exotic Lycopodium tracers (11,300 ± 400 spores) were added to the samples in order 
to calculate overall pollen concentration (Table 1). Pollen concentration values 
(number of pollen grains/gram of material) for each coorolite were calculated by 



m 



number 



times the amount 



from 



mine 



medicines. Modern fecal stndipc havp ehnwn that the more 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 207 



recently a pollen type was ingested the higher concentration value a fecal sample 
will have (Kelso 1976; Williams-Dean 1978). As the time from ingestion of a pollen 
type lengthens, the pollen concentration value for that sample decreases. The 
reason for this decrease is that pollen tends to become caught in the intestinal 
luminal folds and can be excreted many days after initial ingestion. Some pollen 
types can be excreted for up to one month after ingestion (Williams-Dean 1978). 
Presence of high pollen concentration values in coprolites therefore indicate 
that the economic pollen types observed in the samples were ingested recently. 
Concentration values of over 100,000 pollen grains/gram of material usually 
indicate that pollen was recently ingested (Sobolik 1988). Samples which contain 
less than this amount may contain economic pollen types that were intentionally 
ingested many days before the sample was deposited (Sobolik 1988, 1991). Such 
samples will also contain a wide variety of unintentionally ingested, background 
pollen types. Therefore it is harder to recognize intentionally ingested pollen types 
in samples which contain less than 100,000 pollen grains/gram of material. Other 
factors influencing pollen abundance, such as the amount of pollen a plant 
produces and dispersal method, should also be considered in assessing the 
significance of the pollen. 



DISCUSSION 



The prehistoric Caldwell Cave population intentionally ingested at least four 
pollen types (Table 1). These pollen types are considered economic because they 
occur in high frequencies in the samples and most likely do not indicate acciden- 
tal ingestion of pollen through contamination on food, in water supplies, and 
through breathing. The four economic pollen types are grass (Poaceae), Mormon 
tea (Ephedra), mesquite (Prosopis), and cactus (Cactaceae of the non-Opuntia 

type). 

Twelve of the Caldwell Cave coprolites had concentration values of over 
100,000 pollen grains/gram of material; ten samples contained extremely high 
percentages of grass pollen and aggregates; one sample contained a high frequency 
of cactus pollen; one sample contained a high frequency of Mormon tea pollen 
(Table 1). The high concentration values of these samples suggest that grass, 
Morman tea, and cactus flowers or inforescences were most likely intentionally 
ingested only a few days before the samples were deposited. 

Grass pollen is particularly prevalent in the coprolite samples. Grass pollen, 
which is wind-pollinated, was found at high frequencies in 29 of the 30 samples 
which contained a statistically valid pollen count. A majority of the samples with 
abundant grass pollen also contained aggregates of grass pollen, possibly indi- 
cating the ingestion of anthers. The presence of pollen aggregates in non-copro- 
litic samples may, however, indicate a variety of depositional patterns (Gisn 
1991). The one coprolite sample which did not contain a high frequency of grass 
pollen, sample 28, contained a high frequency of Cactaceae pollen and aggregates 
of this type. Grass pollen was observed in samples which had high, intermediate, 
and low concentration values indicating variation in the length of time from 



208 



SOBOLIK and GERICK 



Vol. 12, No. 2 



TABLE 1.— Major pollen frequencies from 



Sample 
Number 



1 
2 
3 
4 

6 
7 

8 

10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
26 
27 
28 
29 
30 
31 
32 



Morph. 
Category 

I 
I 

II/III 
11/ III 
II/III 
II/III 
II/III 
II/III 
II/III 
II/III 
II/III 
II/III 
II/III 
II/III 
II/III 
II/III 
II/III 
IV 
IV 
IV 
IV 
IV 
IV 
IV 
IV 

IV 
IV 
IV 
IV 
IV 



Ephedra Prosopis 

% 



Cactaceae Poaceae Concentra- 



Count 



°/c 



% 



% 



tion Values Totals 





1 







2 



4 














4 



4 
2 


47 

75 


1 
5 

51 





























1 



1 

47 
1 

3 
6 




1 

2 

1 

41 




1 



1 




18 










2 

1 

2 





91 a 


1 
1 



91 a 
98 a 
98 a 
85 a 
45 
37 
20 
24 
80 a 

92 a 
31 
91 a 
98 a 

93 a 
87 a 
89 a 
18 

96 a 
36 
38 
43 
49 
98 a 
23 
12 

2 

86 a 
47 
25 a 
22 



105,805.2 
240,125.0 
283,233.8 
120,860.9 
4,731.5 
15,281.9 
9,746.3 
5,244.8 
77,549.0 
154,433.3 
35,123.7 
80,440.7 
700,376.2 
254,684.6 
110,684.4 
27,775.2 
21,375.8 
96,857.1 
3,550.7 
8,999.6 
9,658.1 
9,783.5 
300,923.9 
67,261.9 
303,277.4 
128,866.4 
113,758.4 
3,705.7 
67,472.5 
96,660.8 



250 
238 
386 
246 
204 

213 

207 

200 

210 

205 

221 

210 

313 

293 

239 

234 

227 

216 

200 

205 

200 

200 

245 

200 

416 

222 

200 

223 

206 

211 



a 



aggregates 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 209 



ingestion of grass pollen to sample deposition. Grass pollen may become ingested 
by eating grass seeds, which, due to their small size, are difficult to separate from 
grass inflorescences. Holloway (1983) also observed high percentages of grass 



samples he analyzed, and Murry 



sam 



Ephedra pollen was found at high frequencies in three of the Caldwell Cave 



samples (sam 



m 



containing a high frequency of Ephedra pollen were in morphological category IV, 
indicating that the coprolites were diarrhetic. This suggests that only people with 



ems ineested Mormon 



em 



Mormon 



medicinal remedies 



Moore 



Moerman 1986). Mormon 



mmer 



Mormon 



(M 



observed 



m 



economic 



sidering that Ephedra is wind-pollinated-its pollen occurs in the environment 
at higher frequencies than insect-pollinated types. Ephedra may be economic in 
this case, however, if it was ingested some time before the sample was deposited 
(Sobolik 1988). Murry (1980) observed Ephedra at high frequencies (92%, 86%) 



in two or the sam 



Cactaceae pollen (of the non-Opuntia variety) was observed at high frequen- 
cies in sample 28 (morphological category IV). This sample also had a high pollen 
concentration value (Table 1). Pollen of Cactaceae, which are insect-pollinated 
plants, was also observed at a high frequency (79.3%) in the colon contents of 
a mummified child, which also contained tiny non-Opuntia cactus seeds (Murry 
1980). Cactaceae pollen was not observed in other coprolites analyzed 



from 



area. 



The coorolite sam 



am of material 



may also be economic. Grass pollen was again observed 



sam 



deposited. The high frequencies of grass pollen presem 
n tvoe was most likely intentionally ingested. Mormon 



observed 



m of material 



most 
Sampl 



(mesquite) pollen but have only intermediate pollen 



me 



210 



SOBOLIK and GERICK Vol. 12, No. 2 



category IV and one in morphological category 11/ III . This indicates that mesquite 
may have been used to help cure diarrhea, although other uses are also possible. 
For example, mesquite leaves are useful as a medicinal tea for stomach ailments 
and to cleanse the system (Neithammer 1974). In the process of preparing 
mesquite leaves for a medicinal tea, pollen could also become incorporated into 
the tea, either intentionally or unintentionally. Mesquite also may have been 



m 



mes 



in other coprolite studies in the area. 



1 ' ' - — — - — — - — — — / ? 

Mesquite pollen was not observed 



CONCLUSION 



Analysis of the pollen content of Caldwell Cave coprolites indicates the 
probable ingestion of Mormon tea (Ephedra) and the possible ingestion of mes- 
quite (Prosopis) pollen in medicinal teas to help cure diarrhea. There is a strong 
correlation of diarrhetic coprolites with high frequencies of Mormon tea pollen; 
one diarrhetic coprolite contained a high frequency of mesquite pollen. 

The addition of the present study of 32 coprolites to the previous study of 
eight coprolites from Caldwell Cave (Hollo way 1983), strengthens the conclusion 
that the prehistoric inhabitants of the area used Ephedra pollen in medicinal teas 
to cure diarrhea. Prosopis and Larrea pollen may also have been used for the same 
purpose. 

Determining prehistoric medicinal plant usage is difficult, since plants were 
also used extensively as foods and for other purposes. The best indications of 
medicinal plant usage are provided through the analysis of coprolites, particularly 
pollen content. Analysis of the Caldwell Cave coprolites demonstrates that it '" 
possible to determine prehistoric medicinal plant use through careful analysis of 
the pollen preserved in coprolites. 



is 



ACKNOWLEDGEMENTS 



We 



for allowing us to analyze the Caldwell Cave coprolites, and Joyce Blasi, Bryan, Texas, 
for her editorial comments. We would also HWp *n fVi^L- v a noK« \a Rnnnf Tr Dpborah 



M. Pearsall, Suzar 
Monty McCrossen 



Laboratory 



LITERATURE CITED 

\N, LOWELL JOHN, and KATHERINE BRYANT, VAUGHN M., JR. 1974. Prehis- 

SIVA SAUBEL. 1972. Temalpakh: Cahu- tone diet in southwest Texas: The copro- 

illa Indian Knowledge and Usage of ht e evidence. American Antiquity 39: 

Plants. Malki Museum Press, Morongo 407-420 

Indian Reservation, Phoenix. BURLAGE, HENRY M. 1968. Index of plants 



Winter 1992 



JOURNAL OF ETHNOBIOLOGY 



211 



LITERATURE CITED (continued) 



of Texas with Reputed Medicinal and 
Poisonous Properties. Austin. 

CURTIN, LEONORA S. M. 1949. By the 
Prophet of the Earth. San Vicente Foun- 
dation, Santa Fe. 

FRY, GARY F. 1985. Analysis of fecal material. 
Pp. 127-154 in The Analysis of Prehis- 
toric Diets. Robert I. Gilbert, Jr. and 
James H. Mielke (editors). Academic 
Press, New York. 

GISH, JANNIFER W. 1991. Current percep- 
tions, recent discoveries, and future 
directions in Hohokam palynology. Kiva 
56:237-254. 

HOLLO WAY, RICHARD G. 1983. Diet and 
medicinal plant usage of a Late Archaic 
population from Culberson County, 
Texas. Bulletin of the Texas Archaeologi- 
cal Society 54:319-329. 

JACKSON, A. T. 1937. Exploration of certain 
sites in Culberson County, Texas. Bulle- 
tin of the Texas Archaeological and Pale- 
ontological Society 9:146-191. 

KELSO, GERALD. 1976. Absolute pollen fre- 
quencies applied to the interpretation of 
human activities in northern Arizona. 
Unpublished Ph.D. dissertation, Depart- 
ment of Anthropology, University of 

Arizona, Tucson. 
MOERMAN, DANIEL E. 1986. Medicinal 

plants of native America. University of 

Michigan, Museum of Anthropology, 

Technical Reports No. 19. 
MOORE, MICHAEL. 1979. Medicinal Plants 

of the Mountain West. Museum of New 

Mexico Press, Santa Fe. 
MURRY, ROBERT E., JR. 1980. Paleo-diet 

at Granado Cave: The fecal evidence. 



Paper presented at the 51st Annual Meet- 
ing of the Texas Archaeological Society, 

Austin. 

NIETHAMMER, CAROLYN. 1974. Ameri- 
can Indian Food and Lore. Macmillan, 
New York. 

REINHARD, KARL J., DONNY L. HAMIL- 
TON, and RICHARD H. HEVLY. 1991. 
Use of pollen concentration in paleo- 
pharmacology: Coprolite evidence of 
medicinal plants. Journal of Ethnobiology 

11:117-132. 

SAYLES, E. B. 1941. Infant burial in carrying 
basket. Bulletin of the Texas Archaeologi- 
cal and Paleontological Society 13:77-87. 

SOBOLIK, KRISTIN D. 1988. The impor- 
tance of pollen concentration values 
from coprolites: An analysis of southwest 
Texas samples. Palynology 12:201-214. 

. 1991. The prehistoric diet and 

subsistence of the lower Pecos region as 
reflected in coprolites from Baker Cave, 
Val Verde County, Texas. University of 
Texas, Texas Archeological Research Lab- 
oratory, Studies in Archeology, No. 8. 

TANNER, R. W. 1949. The Caldwell Ranch 
sites: A distinctive culture complex in 
the northeastern trans-Pecos. Unpub- 
lished Masters thesis, Department of 
Anthropology, The University of Texas, 

Austin. 
WILLIAMS-DEAN, GLENNA J. 1978. Ethno- 
botany and cultural ecology of prehis- 
toric man in southwest Texas. Unpub- 
lished Ph.D. dissertation, Department of 
Biology, Texas A & M University, College 
Station. 



BOOK REVIEW 



Traditional Plant Foods of Canadian Indigenous Peoples: Nutrition, Botany 

and Use. Harriet V. Kuhnlein and Nancy J. Turner Volume 8 in the Food 
and Nutrition in History and Anthropology Series, edited by Solomon Katz. 
New York and Philadelphia: Gordor i and _Br-h ^cie^ P^ishe^ ^ 



maps. $88.00 (hardbound), $38 



members 



most 



America 



'few and far between." It is indeed encouraging that such a highly 



212 



BOOK REVIEW Vol. 12, No. 2 



detailed and thorough report on food plants and nutrition of the Indigenous 
Peoples of Canada has appeared. The two authors— a well known nutritionist 



from McGill 



are to be congrat- 



mas 



Montour, a Native Canadian and a staff member 



the Assembly of First Nations, sets the tone of the book: "We need to work ha 
together to preserve our knowledge and to protect the environments of the pla 
foods of the world's indigenous people. This book is a good step along the way 
In the Acknowledgments, the authors indicate the great number of individuals 
aboriginal consultants, students, governmental agencies, and scientific 



make 



colleagues 

masterpiece that it is. The Introduction, Chapter 1, tells us that "The scientific 
literature was searched for nutrient information of approximately 1,050 species 
that were identified as edible and available in Canada," and that "... nutritional, 



more 



"are 



included. 



There follows Chapter 2, "What's So Special About Indigenous Foods? 
Amongst other "specialties" is the usefulness of information in "genetic research, 
in enhancing existing crops or ... developing new ones." Chapter 3 deals in depth 



with "An Overview 



chapter— a maj 



are considered the 



botany and methods of use of indigenous plant foods of Canada. A compre- 
hensive list of plant food species makes up Chapter 5, a convenient tabular 
summary of earlier chapters. Another tabular chapter deals with "Nutrient Values 
of Traditional Plant Foods"— a most useful addition that occupies 162 pages. 

The Bibliography comprises 519 items and is followed by three Appendices: 
Linguistic Affiliations and Locations of Indigenous Peoples of Canada (with three 
maps); Species by Common Name; and Species by Botanical Name. The Index 
contains both common and scientific plant names, and nutritional and other 
chemical constituents of the food plants. 

This volume will certainly long remain an example of the very finest in 
ethnobotanical literature. Furthermore, its utilitv will be evident as a manual for 



classroom 



instrument 



of ethnobotany. 



Richard Evans Schultes 
Director Emeritus 
Botanical Museum of Harvard 
Cambridge, Massachusetts 



J 



Winter 1992 



RECENT DOCTORAL DISSERTATIONS 
OF INTEREST TO ETHNOBIOLOGISTS: 

FALL 1991-FALL 1992 






TERENCE E. HAYS 

Department of Anthropology and Geography 

Rhode Island College 
Providence, RI 02908 

and 

JOSEPH E. LAFERRIERE 

Arnold Arboretum of Harvard University 

22 Divinity Avenue 
Cambridge, MA 02138 



This is the tenth in an annual series of bibliographies listing selected disser- 



from 



made 



which ones mi 



economic 



Agriculture 



American Studies, Anthropology, Biology, Botany, Chemistry, Ecolc 
Geography, Health Science, Home Economics, Language, Linguistics, 
Paleoecology, Physical Geography, Sociology, and Zoology were considered for 
inclusion in the list. An attempt was made to be as inclusive as possible, but some 



may have been overlooked. Comments 



welcome for items 



Social 



September 



September 1991-August 1992; Volume C (European Dissertations): Fall 



Summer 1992. Note 



the abstracts appear, rather than the dates of acceptance of the dissertations 



m 



The 



major 



the page(s) in D. A. on which the abstract may be found, University Microfilms 
order number, and the ISBN number when this information was included 

Most of the dissertations accepted at institutions in the United States and 
some of those from Australia, Canada, South Africa, and the United Kingdom 
may be obtained from University Microfilms International, P.O Box 1764 Ann 



MI 48106-1346, either on microfilm 



Q 



ures 



varies with the quality of the original. Current prices may be ^£%££* 
800-521-3042; 313-761-4700 from Alaska, Hawaii, or Michigan; or 800-343-52^ from 
Canada. Further information and current prices, as well as information regarding 



214 



HAYS and LAFERRIERE Vol. 12, No. 2 



UMI order number, may 



from UMI Dissertations Information Service, 300 North 
Dr. MI 48106-1346, USA. 



RESUMEN.— En este bibliografia se incluyen disertaciones recientes de interes 
a los etnobiologos. Por cada uno se da el numero de la pagina donde se halla 
el resumen en Dissertation Abstracts (D. A.), y el numero de encargar un ejemplar 
de la disertacion de University Microfilm International, P.O. Box 1764, Ann 
Arbor, MI 48106-1346 USA (telefono: 313-761-4700 o 800-521-3042; desde Canada 
800-343-5299) . 



RESUME.— Cette bibliographie comprend quelques dissertationes recentes 
d'interet aux ethnobiologistes. Chez chaqu-une on donne le numero de la page 
ou se trouve le resume dans Dissertation Abstracts (D.A.), et le numero de com- 



Microfilm 



1346 



3042 



Adeetuk, Thomas Akanpasagi. 1991 . Land tenure and food production in northern 
Ghana: 1900-1985. University of Wisconsin-Madison, 215 pp. D.A. 52(7): 
3507-B. Order no. DA9128892. 

AL-Duleimi, Saadoon J.F. 1990. An analysis of factors that influence adoption 
of improved agricultural practices among Iraqi farmers. University of Keele 
(United Kingdom), 340 pp. D.A. 53(1):196-197-A. Order no. BRDX95704. 

Asfaw, Zemede. 1990. The barley of Ethiopia: A focus on the infraspecific taxa. 
Uppsala Universitet (Sweden), 33 pp. ISBN: 91-554-2454-6. D.A. 53(2):235-C. 

Baied, Carlos Alberto. 1991. Late-Quaternary environments and human occupa- 
tion of the south-central Andes. University of Colorado at Boulder, 151 pp. 
Directors: David L. Green; Vera Markgraf. D.A. 53(2):537-A. Order no. 
DA9220391 . 

Balde, Mamadou Aliou. 1990. Biological and phytochemical investigations on 



medicine [Pavetta 



abyssinica] 



um 



Amazon 



Joseph 



University of Georgia, 150 pp. D.A. 52(4):1823-B. Order no. DA9117276. 



The 



(Bombaceae): A phylogenetic approach. Washington University, 421 pp- 
D.A. 52(10) :5057-B. Order no. DA9209159. 

vis, John. 1990. The effects of land use and burial on the amino acid compo- 
sition of soils. Volumes I and II. University of Reading (United Kingdom), 
676 pp. D.A. 52(6):2187-A. Order no. BRDX93624. 

k, Hans Troster. 1991. The taxonomy and economic botany of the cultivated 
guarana and its wild relatives and the generic limits within the Paullinieae 
(Sapindaceae). Volumes I and II. City University of New York, 551 pp- 
D.A. 52(9):4550-B. Order no. DA9207049. 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 215 



Bellquist, Julia Bonner. 1991. Ecological etymology: The history of the Germanic 

carnivore names in their archaeological, ecological, and linguistic setting. 

University of Texas at Austin, 472 pp. Supervisor: Edgar C. Polome. D.A. 

52(12):4312-A. Order no. DA9212486. 
Bermawie, Nurliani. 1990. Isozymic variability and barriers to hybridisation 

between Capsicum chacoense and two purple-flowered species (Capsicum 

pubescens and Capsicum tovarii). University of Reading (U.K.), 273 pp. D.A. 

52(4):1814-B. Order no. BRDX92874. 
Bian, Ling. 1991. Effects of spatial scale on estimating the relationship between 

vegetation and topography in a mountainous environment. University of 

North Carolina at Chapel Hill, 213 pp. Director: Stephen J. Walsh. D.A. 

52(10):3692-A. Order no. DA9207935. 
Bjorklund, Jeffrey Allen. 1991. Studies on the biosynthesis of cocaine and related 

alkaloids in Erythroxylum coca. University of Minnesota, 261 pp. D.A. 52(3): 

1431-B. Order no. DA9125793. 
Bowen, Jonathan Emerson. 1992. The late prehistory of northwestern Ohio. 



University, 164 pp. Adviser: Richard W 



Order no. DA9218957. 
Brough, Susan Helen. 19 



(Manihot 



Cranz). University of Nottingham 



Order no. BRD-94381. 



Brown, N.D. 1990. Dipterocarp regeneration in tropical rain forest gaps of dif- 
ferent sizes. University of Oxford (U.K.), 184 pp. D.A. 52(8):3971-B. Order 

no. BRD-94146. 
Browne, B.J. 1989. The environmental history of Washing Lough, Kilrea, Co. 

Deny, Northern Ireland. University of Ulster (Northern Ireland), 249 pp. 

D.A. 52(10) :5171-5172-B. Order no. BRDX94802. 
Buhion, Caroline J. 1991. Characterization and 

archaeological wood. Clemson University, 258 pp. D.A. 52(12):6171-B. Order 

no. DA9213073. 
Burford, Mark Derek. 1990. Chemotaxonomic study of feverfew (Chrysanthemum 
parthenium). University of Technology, Loughborough (U.K.), 272 pp. D.A. 

52(1):277-B. Order no. BRDX95690. 
Buriachs i faw<: FranrPQr 1QQD Palinoloeia dels dolmens de l'alt emporda i 



conservation 



dels diposits quaternaris de la Cova de l'Abreda (Serinya, Pla de l'Estany) 
i del Pla de l'Estany (Olot, Garrotxa): Evolucio del paisatge vegetal i del clima 
des de fa mes de 140,000 anys al N.E. de la Peninsula Iberica [Palynology 
of dolmens of the Alt Emporda of the quaternary deposits of the Cova 
de l'Arbreda (Serinya, Pla de l'Estany) and of the Pla de l'Estany (Olot, 
Garrotxa): Evolution of the vegetation and climate from over 140,000 years 
aco in thp nnrthpa^pm nart of the Iberian Peninsula. 1 (Dissertation in 



Autonoma de Barcelona (Spam 



D.A. 52(4):561-C. 

tro, Vilma. 1991. The microclimate of corn 
relationship with Dalbulus maidis, some aphic 
University of Illinois at Urbana-Champaign 
D.A. 52(3):1034-A. Order no. DA9124388. 



216 



HAYS and LAFERRIERE Vol. 12, No. 2 



Charman, Daniel John. 1990. Origins and development of the Flow Country 
blanket mire, northern Scotland, with particular reference to patterned fens. 
University of Southampton (U.K.), 437 pp. D.A. 52(8):3998-B. Order no. 

BRDX94050. 
Claeson, Per. 1990. Pharmacognotic studies of scented myrrh (Commiphora guidotti) 
with emphasis on the biological activities of the isolated sesquiterpene T- 
cadinol. Uppsala Universitet (Sweden), 46 pp. ISBN: 91-554-2653-0. D.A. 

53(2):308-309-C. 

Cole, Roy. 1991. Changes in drought-coping strategies in the Segu region of 
Mali. Michigan State University, 279 pp. Chairperson: Gary Manson. D.A. 
53(1): 258- A. Order no. DA9216291. 

Conard, Nicholas John. 1990. Tonschesberg and its position in the Paleolithic 
prehistory of northern Europe. Yale University, 358 pp. D.A. 52(3): 974- A. 
Order no. DA9122271. 

Cruz, Maria L. 1991. Appropriate technology and shrimp mariculture develop- 
ment in Mexico. Rutgers The State University of New Jersey- New Brunswick, 
320 pp. Director: Bonnie J. McCay. D.A. 52(12):4383-4384-A. Order no. 
DA9213956. 

Cuervo, Alfredo Carabot. 1990. Chemical studies on steroidal sapogenin-pro- 
ducing plants of Venezuela. Council for National Academic Awards (U.K.), 



164 pp. D.A. 52(3):1366-B. Order no. BRDX92193. 



The environmental social movement 



rural 



Frank R. Scarpitti. D.A. 52(9): 3435- A. Order no. DA9206384. 
th, Margaret Redfern. 1982. [sic] Seasonality from shells: Growth-line studies 
of the cockle, Cerastoderma edule L., and the limpet, Patella vulgata L., as a 
guide to prehistoric shell-collecting activities. University of Sheffield (U.K.), 
329 pp. D.A. 52(9):3326-A. Order no. BRDX94502. 



tifolium Nutt. and Salvia apiana Jeps. 1 
52(7):3616-B. Order no. DA9136843. 
>ux, Marie-Genevieve. 1990. Contribut 



from 



[Con- 



tributions to the study of the genus Albizzia.] Dissertation and abstract in 

French.) Universite de Dijon (France), 75 pp. D.A. 52(3):432-C. 
Doherty, Deborah Ann. 1987. Maasai pastoral potential: A study of ranching 

in Narok District, Kenya. McGill University (Canada), 512 pp. ISBN: 0-315- 

64098-7. D.A. 53(1):198-A. Order no. DANN64098. 
Dumont, Clayton Wayne, Jr. 1991. Loggers and radical environmentalists: Cul- 



timber 



Johnson 



Adviser: 



Earwood, Caroline Mary Emmett. 1990. Domestic wooden artifacts from pre 



man 



and use. Volumes I and II. University of Exeter (U.K.), 797 pp. D.A. 52(3): 



974-975- A. Order no. BRD-92410. 



Wisconsin-Madison 



temperate 



pho 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 217 



Enloe, James Gordon. 1991. Subsistence organization in the Upper Paleolithic: 
Carcass refitting and food sharing at Pincevent. University of New Mexico, 
437 pp. D.A. 52(10):3640-A. Order no. DA9208123. 

Erkkila, Daniel Lee. 1991. Valuation of forest-based nonmarket outputs: A cost- 
price approach. University of Minnesota, 176 pp. D.A. 52(8):3972-B. Order 
no. DA9202809. 

Ezzo, Joseph Anthony, Jr. 1991. Dietary change at Grasshopper Pueblo, Arizona: 
The evidence from bone chemistry analysis. University of Wisconsin-Madison, 
411 pp. Supervisor: T. Douglas Price. D.A. 52(11): 3979-3980- A. Order no. 
D A9207559 . 

Farah, Mohamed Hussein. 1991. Isolation and identification of pharmacologically 
active compounds from plants used in Somali traditional medicine. Uppsala 
Universitet (Sweden), 48 pp. ISBN: 91-554-2700-6. D.A. 53(2):315-C. 

Farias- Sanchez, Jose Antonio. 1991. Ecology, culture and utilization of the mussel 
Brachidontes recurvus (Refinesque), in the context of an integrated manage- 
ment approach to Boca del Rfo-Mandinga estuarine system, Veracruz, Mexico. 
University of Stirling (U.K.), 355 pp. D.A. 52(11):5595-B. Order no. BRD- 

95291 . 
Fiddes, Nick. 1989. Meat: A natural symbol. University of Edinburgh (U.K.), 

294 pp. D.A. 52(7): 2607- A. Order no. BRD-93862. 

Finch, John David. 1991. Coffee, development, and inequality in the Papua 
New Guinea highlands. City University of New York, 333 pp. Adviser: 
Mervyn J. Meggitt. D.A. 52(9): 3329-3330- A. Order no. DA9207073. 

Finney, Mark Arnold. 1991. Ecological effects of prescribed and simulated fire 
on the coast redwood (Sequoia sempervirens) (D. Don) Endl.). University of 
California-Berkeley, 194 pp. D.A. 52(8):3972-B. Order no. DA9203560. 

Foller, Maj-Lis Annette. 1990. Environmental changes and human health: A 
study of the Shipibo-Conibo in eastern Peru. Goteborgs Universitet (Sweden), 

296 pp. D.A. 53(1):61-C 
Ghauri, Muhammad Jahangir. 1991. Modeling yield of Dalbergia sissoo (Shisham) 

for irrigated plantations of the Punjab, Pakistan. Michigan State University, 

133 pp. D.A. 52(6):2846-B. Order no. DA9134123. 
Gnabre, Jean-Noel. 1991. Antiallergic activity of Tylphora sylvatica. University of 

Arizona, 198 pp. D.A. 52(3):1367-B. Order no. DA9123477. 
Goland, Carol Ann. 1991. Cultivating diversity: Field scattering as agricultural 

risk management in Cuyo Cuyo, Department of Puno, Peru. Volumes I 

and II. University of Michigan, 585 pp. Chair: Jeffrey R. Parsons. D.A. 

52(7): 2603- A. Order no. DA9135599. 
Graffam, Gray Clayton. 1990. Raised fields without bureaucracy: An archa- 
eological examination of intensive wetland cultivation in the Pampa Koani 
Zone, Lake Titicaca, Bolivia. University of Toronto (Canada), 398 pp. ISBN: 
0-315-59810-7. Chair: J.N. Ingham. D.A. 52(11):3980-A. Order no. DANN 

59810. 
Graham, Margaret Anne. 1991. Dimensions of malnutrition and hunger among 
children in an Andean community. Michigan State University, 265 pp. 
Chair: Ann V. Millard. D.A. 53(1):198-A. Order no. DA9216307. 



218 



HAYS and LAFERRIERE Vol. 12, No. 2 



Phytochemical studies of selected species of 
Piper. University of Iowa, 161 pp. D.a. 52(3):1436-B. Order no. DA9122061. 



Gumerman, George John. 1991. Subsistence and compl 
diverse socio-economic groups at Pacatnamu 



Timothy 



52(12): 4381- A. Order no. DA9213668. 

'os, Mekonen. 1989. Phytochemical and pharm 



Somalian medicinal plant used to treat asthma. Uppsala 

Universitet (Sweden), 53 pp. ISBN: 91-554-2461-9. D.A. 53(2):315-316-C. 

Hammett, Julia Elizabeth. 1991. Ecology of sedentry societies without agriculture: 

Paleoethnobotanical indicators from native California. University of North 

Carolina at Chapel Hill. 274 pp. Director; Richard A. Yarnell. D.A. 52(10): 

3640-3641- A. Order no. DA9207953. 

Hammond, David Scott. 1991. The restoration of tropical dry forest after agri- 
culture in Chiapas, Mexico. University of East Anglia (U.K.), 253 pp. D.A. 
52(6):2846-B. Order no. BRDX93486. 

Hasler, Andreas Rolf. 1990. Flavonoide aus Ginkgo biloba L. und HPLC-Analytik 
von Flavonoiden in verschiedenen Arzneipflanzen. [Flavonoids of Ginkgo 
biloba L. and HPLC analyses of flavonoids in various medicinal plants.] 
(Dissertation in German.) Eidgenossische Technische Hochschule Zurich 

(Switzerland), 306 pp. D.A. 52(4):618-C. 
Higginbotham, James Arnold, III. 1991. Artificial fishponds in Roman Italy during 



Empire. University of Michiga 



John 



no. DA9124020. 



history and paleoclima 



paleolimnological record of Lake Mirag 



University of Florida, 110 pp. D.A. 52(10):5070-B. Order no. DA9209022. 
Hogan, Michael. 1991. Land tenure and differentiation in Nuevo Lima, Peru: 

Articulated development in the high rainforest. University of California, 

Riverside, 307 pp. Chairperson: Michael Kearney. D.A. 52(12):4384-A. 

Order no. DA9214731. 
Hudson, Jean Leslee. 1990. Advancing methods in zooarchaeology: An ethno- 

archaeological study among the Aka. University of California, Santa Barbara. 

381 pp. Chair: Michael Jochim. D.A. 52(5):1795-A. Order no. DA9130092. 
Hueske, Kirby Lyn. 1991. The neurological effects oiSolanum dimidiatum in mice. 



A&M 



The search for bioactive constituents from 



(Annonaceae). Purdue University. 244 pp. D.A. 52(10):5281-B. Order no 

D A9201338 . 
Inman, Alastair. 1990. Foraging decisions: The effects of conspecifics and environ 

mental stochasticity. University of Oxford (U.K.), 210 pp. D.A. 52(8):3999-B 

Order no. BRD-94143. 
Javed, Nasim. 1991. Hydrologic responses to fuelwood harvest and slash dispose 

on a pinyon-juniper dominated grassland site in the Gila National Forest 

New Mexico. New Mexico State University, 148 pp. D.A. 52(8):3985-B 

Order no . D A9201915 . 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 219 



Jimenez-Osornio, Juan Jose-Maria. 1991. Ecological basis of weed management 
in the chinampa agroecosystem. University of California- Riverside, 167 pp. 
D.A. 52(12):6142-B. Order no. DA9214736. 

Jones, Andrew Kenneth George. 1991. The fish remains from Freswick Links, 
Caithness. University of York (U.K.), 396 pp. D.A. 52(7):2604-A. Order 
no. BRDX93985. 

Jones, Cynthia S. 1990. The developmental basis of leaf shape variation in a 
wild and a cultivated subspecies of Cucurbita (C. argyrosperma) . University 
of California-Berkeley, 222 pp. D.A. 52(4):1815-B. Order no. DA9126630. 

Jones, John Glendon. 1991. Pollen evidence of prehistoric forest modification 
and Maya cultivation in Belize. Texas A&M University, 141 pp. Chairman: 
Vaughn M. Bryant, Jr. D.A. 53(1):195-A. Order no. DA9216951. 

Juracek, Kvle Elliot. 1991. GIS and expert system approaches to land suitability 



assessment 



4044 



Kellogg, Douglas Carlton. 1991. Prehistoric landscapes, paleoenvironments, 
and archaeology of western Muscongus Bay, Maine. University of Maine, 
341 pp. Adviser: David Sanger. D.A. 53(1):195-A. Order no. DA9218007. 

Kimsey, Mary Buskin. 1991. A spatial analysis of the causes of tropical defor- 
estation. University of Georgia, 203 pp. Director: Vernon Meentemeyer. 
D.A. 52(9): 3389- A. Order no. DA9206963. 

Kiviat, Erik. 1991. Wetland human ecology. The Union Institute, 187 pp. D.A. 
52(ll):5631-5632-B. Order no. DA9211445. 

Laferriere, Joseph Edward. 1991. Optimal use of ethnobotanical resources by 
the Mountain Pima of Chihuahua, Mexico. University of Arizona, 266 pp. 

D.A. 52(7):3537-B. Order no. DA9136850. 
Lamont, Eric E. 1991. Taxonomy otEupatorium Section Verticillata (Asteraceae). 
City University of New York, 217 pp. D.A. 52(9):4551-B. Order no. DA 



9207091. 



vid Bachenheimer. 1991. Zooarchae 
ly from eastern Massachusetts. Volu 
Major Professor: Mary C. Beaudry 



DA9122923. 



Jean. 1991. Anasazi harvests: Agroclimate, harvest 



Mesa, northeastern Arizona. 

W. Conrad. D.A. 52(9): 



3327- A. Order no. DA9205950. 



Lehtinen, Ari Auskuti. 1991. Northern natures: a study of timber-line conflict 
in Finland. Helsingen Yliopisto (Finland), 112 pp. D.A. 52(4):524-525-C. 

Lennstrom, Heidi Annette. 1992. Intrasite spatial variability and resource utili- 
zation in the prehistoric Peruvian highlands: An exploration of method and 
theory in paleoethnobotany. University of Minnesota, 441 pp. Adviser: 
Christine A. Hastorf. D.A. 53(1):196-A. Order no. DA9215985. 

Li, Minggvang. 1991. The ecology of neotropical forest tree seedlings. University 
of North Dakota, 177 pp. D.A. 52(5):2396-B. Order no. DA9131502. 

Lis, Richard Adam. 1990. A taxonomic revision olHolidiscus Maxim. ( Spira eoideae : 
Rosaceae) based uDon numerical analyses of leaf morphological variation. 



220 



HAYS and LAFERRIERE Vol. 12, No. 2 



University of California-Berkeley, 221 pp. D.A. 52(4):1815-B. Order no. 

DA9126668 . 
Lufumpa, Leyeka Charles. 1991. An economic analysis of agroforestry farming 
systems in Zambia: Application of risk programming and risk-free modelling 
techniques. Iowa State University, 287 pp. D.A. 52(8):3972-B. Order no. 

D A9202374 . 
Magee, Pennie Lou. 1990. "The water is our land:" Peasants of the river Tocan- 
tins, Brazilian Amazonia. University of Florida, 161 pp. Chair: Marianne 
Schmink. D.A. 52(3):980-A. Order no. DA9121625. 
Maitima, Joseph Mworia. 1991. Vegetational environments of later Stone Age 
and early Iron Age cultures in western Kenya. Duke University, 311 pp. 
D.A. 52(8):4097-B. Order no. DA9202500. 

Mansberger, Joe Robert. 1991. Ban yatra: A bio-cultural survey of sacred forests 
in Kathmandu Valley. University of Hawaii, 343 pp. Chairman: Brian Murton. 
D.A. 52(5): 1858- A. Order no. DA9129689. 

Marean, Curtis William. 1990. Late Quaternary paleoenvironments and faunal 
exploitation in East Africa. University of California-Berkeley, 518 pp. D.A. 
52(4):1404-A. Order no. DA9126689. 

McCartney, Peter H. 1989. Paleoeskimo subsistence and settlement in the High 
Arctic. University of Calgary (Canada), 347 pp. ISBN: 0-315-61748-9. Super- 
visor: James W. Helmer. D.A. 52(10):3641-A. Order no. DANN61748. 

Mena, Manuel Francisco. 1991. Prehistoric resource space and settlement at 
the Rio Ibanez Valley (central Patagonian Andes). University of California, 
Los Angeles, 195 pp. Chair: Clement Meighan. D.A. 52(7): 2604- A. Order 
no. DA9200909. 

Menegoni, Lorenza A. 1990. Tuberculosis and health care in Highland Chiapas, 
Mexico: An ethnographic study. New School for Social Research, 365 pp- 
D.A. 52(3):1379-B. Order no. DA9123623. 

Menelaou, Marios Andrea. 1990. Structural and biosynthetic studies of natural 
products of the Asteraceae and Lamiaceae (Solidago pauciflosculosa and Cala- 
mintha ashei). Louisiana State University and Agricultural and Mechanical 
College, 289 pp. D.A. 52(3):1441-1442-B. Order no. DA9123221. 

Merrick, Laura Channing. 1991. Systematics, evolution, and ethnobotany of a 
domesticated squash, Cucurbita argyrosperma. Cornell University, 340 pp- 
D.A. 52(9):4552-B. Order no. DA9204093. 

Mohamad, Aminuddin Bin. 1990. Ecology and silviculture of Calamus manan in 
peninsular Malaysia. University College of North Wales, Bangor (U.K.), 
276 pp. D.A. 52(11) :5606-B. Order no. BRDX95205. 

Morgado, Luiz Balbino. 1991. Nitrogen relationships in maize-beans intercrop- 
ping. University of East Anglia (U.K.), 262 pp. D.A. 53(1):37-B. Order no. 
BRDX95610. 

Morgan, R. Grace. 1991. Beaver ecology / beaver mythology. University of Alberta 
(Canada), 95 pp. ISBN: 0-315-66754-0. Supervisors: H.T. Lewis; R. Gruhn. 
D.A. 53(2):543-A. Order no. DANN66754. 

Naaranlahti, Toivo. 1991. Isolation and analysis of Catharanthus alkaloids with 
special reference to 3', 4'-anhydrovinblastine biosynthesis. Kuopion Yliopisto 
(Finland), 63 pp. ISBN: 951-780-778-3. D.A. 53(2):340-C. 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 221 



Natadecha, Poranee. 1991. Nature and culture in Thailand: The implementation 
of cultural ecology in environmental education through the application of 
behavioral sociology. University of Hawaii, 197 pp. Chairperson: Royal T. 
Fruehling. D.A. 52(5): 1798-1799- A. Order no. DA9129697. 

Newman, Margaret Evelyn. 1990. The hidden evidence from Hidden Cave, 
Nevada: An application of immunological techniques to the analysis of 
archaeological materials. University of Toronto (Canada), 168 pp. ISBN: 
0-315-65865-7. Supervisors: M.A. Latta. D.A. 53(2):540-A. Order no. DANN 

65865 . 
Newnham, Rewi Munro. 1991. Late Quaternary palynological investigations into 
the history of vegetation and climate in northern New Zealand. University 
of Auckland (New Zealand), 374 pp. D.A. 52(8):4098-B. Order no. DA9135932. 



within 



assem 



omic 






52(12):4388-4389-A. Order no. BRDX95528. 
Noblick, Larry Ronald. 1991. The Indigenous palms of the state of Bahia, Brazil. 
University of Illinois at Chicago, 541 pp. D.A. 52(8):3990-B. Order no. DA 

9203385 . 
Norr, Lynette Caryl. 1991. Nutritional consequences of prehistoric subsistence 
strategies in lower Central America. University of Illinois at Urbana-Cham- 
paign, 332 pp. Adviser: David L. Grove. D.A. 52(3):976-A. Order no. DA 

9124465. 
Nshimo, Charles Musa. 1991. Phytochemical and biological studies on Muntingia 

calabura L. University of Illinois-Chicago, Health Sciences Center, 139 pp. 

D.A. 52(5):2518-B. Order no. DA9132108. 
Old, Richard Robert. 1990. W.E.E.D.S.: Western Expert Educational Diagnostic 

System: A computer aided expert system for plant identification. University 

of Idaho, 109 pp. D.A. 52(7):3351-B. Order no. DA9135960. 
Omar, Hishamuddin B. 1989. An assessment of potential of red seaweed Palmaria 



149 pp. D.A. 



palmata for mariculture in the Irish Sea. University of Liv 
183 pp. D.A. 52(11):5609-B. Order no. BRDX95218. 
Ouendeba, Botorou. 1991. Diversity, combining ability, and het 
among African pearl millet landraces. Purdue University, 

53(1 ): 7- B. Order no. DA9215639. 
Parfitt, Bruce D. 1991. Biosystematics of the Opuntia polyacantha complex (Cacta- 
ceae) of western North America. Arizona State University, 123 pp. D.A. 

52(6):2862-B. Order no. DA9134882. 
Payne, Lori Denise. 1991. The alkaloids of Eiythrina: clonal evaluation and 
metabolic fate. Louisiana State University and Agricultural and Mechanical 



College, 175 pp. D.A. 53(2):816-B. Order no. DA9219568. 
lips, Mary T. 1991. Constructing laboratory animals: An etl 



W 






Goffman 



Pharm 



plant Ipomoea pes-caprae (L.) R. Br. (Pak Bung Ta Lae). Uppsala Universitet 
(Sweden), 43 pp. ISBN: 91-554-2656-5. D.A. 53(2):317-C. 



222 



HAYS and LAFERRIERE Vol. 12, No. 2 



Price, Norman William. 1990. The tropical mixed garden in Costa Rica: A potential 
focus for agroforestry research? University of British Columbia (Canada), 
423 pp. ISBN: 0-315-59529-9. D.A. 52(10) :5071-5072-B. Order no. DANN 

59529. 

Ragone, Carol Diane. 1991. Collection, establishment, and evaluation of a germ- 
plasm collection of Pacific Island breadfruit. University of Hawaii, 204 pp. 
D.A. 52(12) :6162-B. Order no. DA9215034. 

Rakotonomenjanahary, Rivo Andriavololonavalona. 1991. Apis mellifica et son 
utilisation a doses homeopathiques. [Apis mellifica and its use in homeo- 
pathic doses.] (Dissertation and abstract in French.) Universite de Dijon 
(France), 79 pp. D.A. 52(3):435-C. 

Rancy, Alceu. 1991. Pleistocene mammals and paleoecology of the western 
Amazon. University of Florida, 164 pp. D.A. 52(8):4097-B. Order no. 
DA9202046. 

Razzouk, Talal Ahmad. 1990. A study of the nutritional adoption of innovations 
by Syrian farmers. University of Nottingham (U.K.), 475 pp. D.A. 52(9): 
4525-B. Order no. BRD-94395. 

Rice, Robert Armstrong. 1990. Transforming agriculture: The case of coffee 
leaf rust and coffee renovation in southern Nicaragua. University of California, 
Berkeley, 349 pp. D.A. 52(4): 1474-1475- A. Order no. DA9126749. 

Rosaldo-May, Francisco Javier. 1991. Ecological role of wild mustard (Brassica 
kaber (DC.) L.C. Wheeler) in the management of soil- pathogenic fungi and 



nematodes in a corn agroecosystem. University of 
155 pp. D.A. 52(12) :6194-B. Order no. DA9206671. 



The Middle Preceramic 



habitation of Nanchoc, northern Peru. University of Kentucky, 721 pp 



Tom / . % „ / .„. _. w _ w _. m ,„„. 

iss, Jon Lester. 1991. Radiocarbon dating of prehistoric rock painting 

A&M University, 231 pp. D.A. 53(1):239-B. Order no. DA9217015 
rpaki, Analya Anastasia. 1987. The paleoethnobotany of the West 

Akrotiri, Thera: A case study. Volumes I and II. University of 5 

(U.K.), 468 pp. D.A. 52(9):3327-A. Order no. BRDX94401. 
unders, Richard M.K. 1990. The systematics of the aquatic fern gem 

Lam., with particular reference to section Rhizosperma (Mey.) Mett. 



Texas 



Academic 



BRDX93188. 



Schmerzler, Seth Martin. 1991. Attitudes toward environmental 
in the Kingdom of Tonga: Observed behavior and implication 
mental education. University of California, Santa Barbara, 
52(10): 3644- A. Order no. DA9208369. 

Schweithelm, James. 1991. Rapid Watershed Assessment for Oute 



Kalimantan case study. University of Hawaii, 282 pp 
n J. Murton. D.A. 52(9):3390-A. Order no. DA920587* 
May. 1991. "Such diet, as befitted his station as clerl 



// 



The 



52(4):1404-A. Order no. DA9127754. 



Minnesota, 332 pp. Adviser: Janet 



Michil 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 223 



Sharma, Uday Raj. 1991. Park-people interactions in Royal Chitwan National 
Park, Nepal. University of Arizona, 275 pp. D.A. 52(10):5039-B. Order no. 
DA9210308. 

Shaw, Leslie Carol. 1991. The articulation of social inequality and faunal resource 
use in the Preclassic community of Colha, northern Belize. University of 
Massachusetts, 424 pp. Director: H. Martin Wobst. D.A. 52(6):2189-A. 
Order no. DA9132913. 

Short, Kevin MacEwen. 1991. Resource management and socioeconomic develop- 
ment in the Japanese coastal fishing industry. Stanford University, 334 pp. 
D.A. 52(9): 3332- A. Order no. DA9206852. 

Silver, Annette Louise. 1991. The Abbott Interaction Sphere. A consideration 
of the Middle Woodland period in coastal New York and a proposal for a 
Middle Woodland exchange system. New York University, 478 pp. Adviser: 
Howard D. Winters. D.A. 52(3): 977- A. Order no. DA9124764. 

Smith, Greg Charles. 1991. Heard it through the grapevine: Andean and Euro- 
pean contributions to Spanish Colonial culture and viticulture in Moquegua, 
Peru. University of Florida, 384 pp. Chair: Prudence M. Rice. D.A. 53(2): 

540- A. Order no. DA9216244. 
Snider, Mary Carol. 1991. Les metaphores animales dans les atlas linguistiques 
et ethnographiques de la France. (Dissertation in French; abstract in English.) 
Brown University, 232 pp. C..\. 52(9): 3268- A. Order no. DA9204958. 



Sobolik, Kristin Dee. 1991. Paleonutrition of the Lower Pecos Region of the 
Chihuahuan Desert. Texas A&M University, 322 pp. Co-Chairs: Vaughn M. 
Bryant, Jr.,; D. Gentry Steele. D.A. 52(9):3327-A. Order no. DA9206563. 

Soedjito, Herwasono. 1990. Root systems of successional and old-growth forest 
species and its role on nutrient dynamics within a tropical rainforest in 
Indonesia. Rutgers The State University of New Jersey, 184 pp. D.A. 52(3): 

1199-B. Order no. DA9123314. 
Song, Cheunsoon Ahn. 1991. Variations in fiber morphology of prehistoric 

textiles from the Seip Group of Mounds: A model for explanation. Ohio 

State University, 325 pp. Co-Advisers: Lucy R. Sibley; Kathryn A. Jakes. 

D.A. 52(5) .1795- A. Order no. DA9130561. 
Thorne, Aida. 1990. Patterns of naming in plant and animal names in Afrikaans. 

University of Pretoria (South Africa). D.A. 52(4):1313-A. 
Uhr, David Vaughn. 1991. Adaptation of tropical maize to temperate zones. 

North Carolina State University, 152 pp. D.A. 52(8):3985-B. Order no. DA 

9203177. 
Van West, Carla Rebecca. 1990. Modeling prehistoric climatic variability and 
agricultural production in southwestern Colorado: A GIS approach. Wash- 
ington State University, 123 pp. Chair: Timothy A. Kohler. D.A. 52(5):1796-A. 

Order no. DA9131108. 
Vincent, Robert Montgomery. 1991. Biological diversity and Third World develop- 
ment: A study of the transformation of an ecological concept into natural 
resource policy. Oregon State University, 169 pp. D.A. 52(5):1859-A. Order 

no. DA9130925. , 4 . .. . . 

Von Gernet, Alexander D. 1988. The transculturation of the Amerindian pipe/ 

tobacco/smoking complex and its impact on the intellectual boundaries 






224 



HAYS and LAFERRIERE Vol. 12, No. 2 



between "savagery" and "civilization," 1535-1935. McGill University 
(Canada), 780 pp. ISBN: 0-315-64092-8. D.A. 53(1):203-A. Order no. DANN 

64092. 
Wahyuono, Sabagus. 1991. Potential anti- infective agents isolated from Artemesia 

pacifica Nutt. and Guardiola platphylla Gray (fam. Asteraceae). University 

of Arizona, 170 pp. D.A. 52(3):1376-B. Order no. DA9123159. 
Watlington-Linares, Francisco. 1990. Adaptive viticulture in the Caribbean basin. 

University of Florida, 195 pp. Chairman: Cesar Caviedes. D.A. 52(3):1035-A. 

Order no. DA9121679. 
Weyer, Edward Moffat, Jr. 1930. [sic] The Eskimos: A study in adaptation to 
environment. Yale University, 497 pp. D.A. 52(3):982-A. Order no. DA 

9115918. 

Wiener, John Dixon. 1990. A social science basis for prescription of land tenture 
for Alaska Natives. University of Colorado at Boulder, 296 pp. Director: 
Risa I. Palm. D.A. 52(3):1035-A. Order no. DA9122658. 

Williams, David Edison. 1991. Peanuts and peanut farmers of the Rio Beni: 
Traditional crop genetic resource management in the Bolivian Amazon. City 
University of New York, 190 pp. D.A. 52(9): 4554- B. Order no. DA9207138. 

Wilson, Curtis James. 1991. An indigenous nineteenth century settlement/sub- 
sistence system in the area of Demarcation Bay, Alaska. State University 
of New York at Binghamton, 401 pp. D.A. 52(4): 1404-1405- A. Order no. 
DA9127180. 

Woo, David. 1991. A feasibility study of using Landsat MSS data to map tropical 
rice fields in Ifugao, Philippines. University of California, Santa Barbara, 
221 pp. Chairman: John E. Estes. D.A. 52(8):3031-A. Order no. DA9204623. 

Zhang, Ke. 1991. Building an expert system based on a geographic information 
system: An example of landuse managment, Inner Mongolia. Texas A&M 
University, 168 pp. Chair: John R. Giardino. D.A. 53(1):260-A. Order no. 
DA9217058. 



BOOK REVIEW 



Psychedelics Encyclopedia. Peter Stafford. Berkeley, CA: Ronin Publishing, 
Inc. (Box 1035, Berkeley, CA 94701), 1992. (Third edition). Pp. iii, 91, and 
420. $24.95. ISBN 0-914171-51-8. 



In this significantly expanded third edition of Psychedelics Encyclopedia, Peter 
Stafford has offered much additional and interesting material. The earlier edi- 
tions were excellent sources of information much of which were difficult to find 
elsewhere, but this third edition provides an incredible amount of new material 
as well as what the earlier editions offered. Among the additions may be cited 
an update of scientific research, particularly chemical research, a discussion of 
changing social and political considerations concerning psychoactive substances, 
an obituary of leaders in the study of psychoactive plants and materials and other 
aspects bringing the story up-to-date. 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 225 



Those who have used and benefited from Stafford's earlier editions will find 
this third one a welcome source book of a fast changing social, political, and scien- 
tific topic. This edition includes material from the biological, psychological, 
physiological, medical, cultural, and sociological aspects of psychoactive drugs 
and their use. Its outstanding characteristic lies in its ability to strike a balanced 
emphasis on the popular and the scientific or technical points of view on psycho- 
active elements, without prejudiced exaggerations. 

At the price asked, the amount of information in this volume is indeed a 



bargain. 



Richard Evans Schultes 

Botanical Museum of Harvard University 

Cambridge, MA 02138 



BOOK REVIEW 



Farmers, Hunters and Colonists: Interaction between the Southwest and 
Southern Plains. Katherine A. Spielmann, editor. Tucson: The University 
of Arizona Press (Conference Proceedings), 1991. Pp. xii, 220. 16 figures; 
10 tables. $35.00 U.S. (clothbound). ISBN 0-8165-1224-8. 



The 



Mexico 



history of modern New Mexico, western 



Oklahoma 



ntury 



The 



which had strikingly different, though interrelated subsistence and settlement 
systems. Simply put, Puebloan farmers traded surplus maize and some manufac- 



more mobile 



areas 



such trade attractive and possible. The authors go beyond this, however, to 
examine whether one of these groups had a social or economic advantage over 



lm 



relationships and exotic goods, and how deeply the availability of traded food 



economic 



emphasizes climatic 



differences between the two areas, and these arguments are dealt with here in 



J. Speth, Spielm 



(mutualism, parasitism 



showing the extent to which the authors perceive sharp edges between these 



groups. 



lm 



(J. Habicht-Mauche, C. Lintz, T. Baugh, D. Wilcox) 
il and historical influences on the region, noting th 
r,Hp reworks, the migration of Athapaskan groi 



226 



BOOK REVIEW Vol. 12, No. 2 



area, the social meaning of exotic goods for ritual and status displays, and even 
the potential for world systems-style core and buffer zones to explain the dynamics 

of these relationships. 

These approaches are not exclusive, as seen particularly in the Speth and 
Wilcox papers. The internal references within this set of papers demonstrate the 
productive possibilities of continuing dialogue between cultural ecologists and 
postprocessualists, who place less emphasis on environmental variables in 
explaining cultural change. The focus on nonmarket, nonhierarchical economic 
systems in this book provides an important contribution to economic anthropology 
beyond the Plains and Southwest. Individual chapters, the summary chapter (by 
Spielmann), index, and bibliography were all carefully prepared, making this book 
much more than a typical volume of conference proceedings. 



Katherine M. Moore 
Department of Behavioral Science 

Bentley College 
Waltham, MA 02154 



/. Ethnobiol. 12(2):227-231 



Winter 1992 



SHORT COMMUNICATION 



ETHNOBOTANY, DISTRIBUTION, AND CONSERVATION 

STATUS OF Ticondendron incognitum IN 
NORTHERN OAXACA, MEXICO 



GARY J. MARTIN 

Department of Anthropology 

University of California 

Berkeley, CA 94720 

and 

SERGIO MADRID 
Estudios Rurales y Asesoria, A.C 
Escuela Naval Militar No. 420 

Colonia Re forma 
Oaxaca, Oaxaca, Mexico 68050 






The description of new species and genera of plants is the bread-and-butter 
work of tropical plant taxonomists, but finding a new family is a rare event. 
Recently, two Costa Rican botanists named a new species of cloud forest tree, 
Ticodendron incognitum Gomez-Laurito and Gomez P., the sole member of a new 
genus and a new family, Ticondendraceae (Gomez-Laurito and Gomez P. 1989, 
1991), placed in the Fagales (Hammel and Burger 1991). The tree had puzzled 
botanists from Mexico to Costa Rica for many years, but evidence from forestry 
and ethnobotanical research in Oaxaca, Mexico suggests that local people of the 
Sierra Norte had discovered its identity, utility, and local distribution long ago. 

The data presented here are derived from two independent studies carried 
out between 1985 and 1992 in the northern Sierra of Oaxaca. One of us (GJM), 
aided by local collectors, conducted an inventory of useful plants in indigenous 
communities of the Sierra Norte. Ethnobotanical data were recorded for more 
than 5,000 specimens of plants in seasonal evergreen, cloud, pine-oak, and tropical 
deciduous forest (de Avila and Martin 1990; Martin and de Avila 1990). Most of 
the collections were made in a Mixe-speaking community, Totontepec, and in 
a Chinantec-speaking community, Santiago Comaltepec. 

The other co-author (SM) coordinated an inventory of timber species in the 
cloud forests of Santiago Comaltepec as part of a larger forestry project in the 
state of Oaxaca. Working with local Chinantec-speakers, he measured the distri- 
bution, density, height, and diameter of trees along a transect from 500 to 2,500 
meters, sampling a total of 210 sites. For purposes of analysis, this altitudinal 
range was split into four strata: (1) from 500-1,000 meters above sea level; 
(2) 1,001-1,500 meters; (3) 1,501-2,000 meters; (4) 2,001-2,500 meters. The 
sampling sites were spread more or less evenly across these strata. 



228 



MARTIN and MADRID Vol. 12, No. 2 



DISTRIBUTION 



Comaltepe 



meters 



meters. This agrees with the altitudinal limits 
)eraohical ranee: accordine to Hammel and 



from 500 to 2,400 meters, beine most common 



and 1,500 meters (Hammel and Burger 1991:89, 91). 

Of the total of 210 sites, Ticodendron was detected in 102, distributed in the 



manner among 



more than 30 eenera measure 



confirming 



most 



The trees were separated into three size classes, roughly correlated with age. 
Size classes were based on the diameter at breast height (dbh): small (young) 
individuals with a dbh of 10-20 cm.; medium (middle-aged) individuals of 20- 
30 cm.; large (adult) individuals of more than 30 cm. The average density, height, 
and diameter of trees in these classes are given in Table 1. 

In their original description, drawn from Costa Rican and Panamanian 
collections, Gomez-Laurito and Gomez P. gave the range of tree height as 7-20 
meters, and the diameter range as 0.40-0.80 meters dbh. The average heights 
observed in the Comaltepec cloud forest fall within this range, but the average 
diameters are less than half the Central American dimensions, even though the 
Oaxaca measurements were taken at 1.1 meters, a slightly lower level than most 
dbh evaluations. 



TABLE 1. 



diameter 



Comaltepe 



Average Average Average 

Size class (Age) density (trees height diameter 

per hectare) in meters in meters 



Small (Young) 4.9 n2 0.11 

Medium (Middle-aged) 6.5 15.6 0.21 

Large (Adult) 17.3 2 0.6 0.46 



Population Average 28.7 17.9 0.35 



ETHNOBOTANY 



demonstrate the imoortance of maki 



specimens when no fertile material 



mate 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 229 



of the tree in Comaltepec and Totontepec, and had puzzled over its identity. After 
recognizing the species in the 1989 publication by Gomez-Laurito and Gomez 
P., we encouraged local collectors to search for fertile material. Some of the 



specimens 



more data on the little known Mexican 



this species. 

From these collections and from further interviewing, we learned that the 
Mixe name is almendras kup, (almond tree), and the Chinantec name 'ma 1 'u' H 



mention 



Mixe 



(Combretaceae) 



Chinantec, rosaceous fruits such as peaches and cherries. One of the Central 
American common names given by Gomez-Laurito and Gomez P. is duraznillo 
(little peach), which is reminiscent of the Chinantec name. 

The Mixe use the tree trunks as roofing beams (horcones) for house con- 
struction. Both the Mixe and the Chinantec consider it an excellent firewood; 
it is said to burn like oak (Quercus) wood, one of the most esteemed fuels. The 
Chinantec note that the fruits are eaten by squirrels, and large numbers of the 
gnawed endocarps can be found on the forest floor. 



CONSERVATION STATUS 



On a walk through the cloud forest near the community of La Esperanza, 
Comaltepec, many felled trees of Ticodendron can be observed. The path is covered 
with fresh leaves, evidence of the firewood cutters who have worked in the area. 

i* 1 * 1 1 



must 



to find trees; they have now arrived in the high-density zone between 1,100 and 



meters 



Although the current effect of firewood cutting on Ticodendron populations 



minimal 



of Oaxaca could lead to increased exploitation in the future. In the markets of 

,- ^ * _.-.- r ««i, /ni,/7vnic cnn Facrarpap^ and 



(Quercus 



wood (Martin 1992). 
Much'o7 mrw^y^harve^ by impoverished Mixtec speakers from moun- 
tain villages to the west of the valley, and the local supplies are rapidly disap- 

^^ _ .- - 1 ,. r r^ i J~ \*s*^r^*r*sc\ rJorilpfpn what 

pearing. If easily accessible populations ot n 
will be chosen as an alternative fuel source? 



become 



The Chinantec of the Sierra Norte, blessed with communal lands tha 

r ^„i :„i „~~^ K^ro nrnrlnrpd and marketed 



Will 



from 



sim 



Many valuable hardwoods such as cedros (various spec.es of Mehaceae) and 
aguacatillos (various species of Lauraceae) would bring a h.gher pr.ce ^ t.rnber 
but machinery and technology are lacking to pull the massive trunks u 
precipitous slopes. Firewood brings a lower price, but transport ,s relattvely 



230 



MARTIN and MADRID Vol. 12, No. 2 



Apart from Ticodendron, the cloud forest trees most heavily harvested for 
firewood correspond to two species that were found to be relatively abundant 
in the forest inventory: Pseudalmedia oxyphyllaria J.D. Sm. (Moraceae), and 
Calyptranthes chytraculia L. var. americana McVaugh (Myrtaceae). Various species 
of Quercus, from higher up on the mountain, are also prized. Unlike some 
shrubby species in the temperate dry forest that are used as a source of fuel, these 
trees branch only near the crown. This morphology requires that the whole tree 

by chopped down when harvesting the wood, rather than selective pruning of 
branches. 

The Chinantec are relatively unconcerned about the felling of these trees for 
firewood, and perhaps they have good reason. They have observed that, because 
of the quantity of the fruit fall, gaps are quickly filled by saplings, and the popula- 
tion regenerates. Yet direct harvesting is not the only danger. Slash-and-burn 



planting 



amount of primary 



om 



dary 



emergent seedlings. When 



primary 



SPECIMENS 



Hammel and Burger (1991:92) cited four specimens from Oaxaca, from three 
different localities. In the expanded list that follows, a number of new collections 



from 



from Index Herbariorum 



The condition of each specimen 



material 



MEXICO, OAXACA: Municipio de San Miguel Chimalapa, Cerro Salomon 
1,850 m, 21 Aug. 1986 (fr), Wendt et al. 5380 (MO); Municipio de Santiagc 
Comaltepec, La Esperanza, 1,600 m, 30 Sep. 1987 (st), Lopez Luna 49 (MEXU. 
MO, UC); Municipio de Santiago Comaltepec, La Esperanza, 1,600 m, 8 Jul. 199( 



Comaltepec 



MEXU, MO, NY, TEX, UC); Municipio 



old fruit collected beneath 



MEXU, MICH, MO, NY 



Municipio de Totontepec, 2 km SW de Totontepec, 1,900 m, 17 June 1986 
Torres & Tellez 8620; Municipio de Totontepec, 1,900 m, 8 Sev. 1986 (st), R 
Reyes 440 (MEXU, UC); Municipio de Totontepec 1 900 m 



(MEXU 



Municipio de Totontepec, 1,900 m, 11 Mar 



Notes 



Thomas Wendt 



of 



~. ~w-««.„„ ^„ lt uiuvcisuy, nds, worxea extensively in Chimalapas, an area ui 
seasonal evergreen, cloud, and pine-oak forest situated in the northeastern 
corner of Oaxaca state near Chiapas. Working with Zoque-speaking collectors, 
he focused on the woody vegetation of this previously little known zone, which 
includes one of the last reserves of tropical forest vegetation in Mexico. The 



Winter 1992 



JOURNAL OF ETHNOBIOLOGY 



231 



Zoque collaborators had seen Ticodendron previously, but did not know if it 
had a local name, or if it was used in the community. These same collectors were 
able to give Zoque names to the majority of forest tree species they observed. 

Ricardo Lopez Luna and Jose Rivera Reyes worked as collectors in the 
ethnobotanical survey supervised by Gary Martin. Ricardo Lopez L. also parti- 
cipated in the forestry inventory carried out by Sergio Madrid. They are con- 
tinuing to make voucher specimens of useful plants in their communities, and 
are seeking flowering material of Ticodendron. 

Rafael Torres and Oswaldo Tellez are botanists from the National Herbarium 
of Mexico, housed in the Instituto de Biologia of the Universidad Nacional 
Autonoma de Mexico. During their extensive collecting in the state of Oaxaca, 
they discovered fruiting material of Ticodendron in the Sierra Norte. 



NOTES 



1 



Superscript letters refer to Low, Medium, and High tones; additional detail on Chinantec and 
Mixe transcription can be found in Anderson (1989) and Schoenhals and Schoenhals (1965). 



ACKNOWLEDGEMENTS 



The forestry inventory was supported primarily by the Ford Foundation. The 
ethnobotanical survey was funded by grants from the Garden Club of America, National 
Science Foundation, Wenner-Gren Foundation, and World Wide Fund for Nature (US). 
Gary Martin was supported during his tenure in Mexico by a Fulbright-Hays graduate 
training award and a fellowship from the Inter-American Foundation. 



LITERATURE CITED 



ANDERSON, JUDILYNN. 1989. Comaltepec 
Chinantec syntax: Studies in Chinantec 
languages 3. Summer Institute of Linguis- 
tics and The University of Texas at Arling- 
ton Publications in Linguistics 89, Arling- 
ton, Texas. 

DE A VILA, ALEJANDRO and GARY J. MAR- 
TIN. 1990. Estudios Etnobotanicos en 
Oaxaca. In Recursos Naturales, Tecnica 
y Cultura: Estudios y Experiencias para 
un Desarrollo Alternativo. Enrique Leff 
(editor). Centro de Investigaciones Inter- 
disciplinarias en Humanidades— Univer- 
sidad National Autonoma de Mexico, 
m Mexico City. 

GOMEZ- LAURITO, JORGE and LUIS D. 
GOMEZ P. 1989. Ticodendron: A new tree 
from Central America. Annals of the 
Missouri Botanical Garden 76:1148-1151. 
1991. Ticodendraceae: A new 



family of flowering plants. Annals of 
the Missouri Botanical Garden 78:87-88. 



WILLIAM 
oak nor a 



The 



Annals of the Missouri Botanical Garden 
78:89-95. Index Herbariorum. 



MARTIN 



in peasant marketplaces. Pp. 212-223 
In Sustainable Harvesting and Marketing 
of Rain Forest Products. Mark Plotkin and 
Lisa Famolare (editors). Island Press, 
Washington, D.C. 

MARTIN, GARY J. and ALEJANDRO DE 
AVILA. 1990. Exploring the Cloud Forests 
of Oaxaca. Gland Switzerland: World 
Wide Fund for Nature Reports, Octo- 
ber/November/December issue. 

SCHOENHALS, ALVIN and LOUISE C 
SCHOENHALS. 1965. Vocabulario Mixe 
de Totontepec. Serie de Vocabularios 
Indigenas Mariano Silva y Aceves No. 14. 
Summer Institute of Linguistics, Mexico, 

D.F. 



232 



BOOK REVIEW Vol. 12, No. 2 



BOOK REVIEW 



Amazon— the Flooded Forest. Michale Goulding. London: BBC Books, and 
New York: Sterling Publishing Co. (387 Park Avenue, South, New York, 
NY 10016), 1989. Pp. 208. $24.95 U.S. (15.00 pounds sterling). ISBN 0-8069- 
7476-1. 



Written by a young man with 12 years of experience in fieldwork in the 
Amazon, including studies in 25 rivers, this book must be recommended to 
all who have no first-hand knowledge of the region but who are desirous of 
learning about it without the usual fanfare that encumbers much of the popular 
and semipopular literature that is flooding our markets today. In addition to 
being aimed at the educated popular market, there is much in the book that 
will be of interest to scientifically oriented students, scholars, and even seasoned 
scientists. 

The book is divided into 10 chapters: (1) Time on the Amazon; (2) Floods; 
(3) Flight in the Forest; (4) Feet in the Forest; (5) Rafts of Life; (6) Lakes; (7) 
Pyramids of Predation; (8) Beaches and Banks, (9) Lower Amazon; and (10) 



There 



lm 



small in number 



ram 



time, m 



picture against biological and geological backgrounds. 
Would that we could have more such intuitivenes 
the great region of the Amazon. 



Richard Evans Schultes 

Botanical Museum of Harvard University 

Cambridge, MA 02138 



BOOK REVIEW 



Amazonia 



Medicine Men 



Raffauf. Oracle, AZ: 



24-7. 



$22 



Similar in format to the masterpiece photographic essay Where the Gods Reign: 
Plants and People of the Colombian Amazon (Schultes 1988), Vine of the Soul is in 
many ways a companion volume. Almost none of the 160 impressive 



from my 



Where the Gods Reign. A q« ote 



// 



The numer- 



(medicine men), and the plants 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 233 



rivers, waterfalls, and other scenes of natural splendor are appropriate in this 
context because medicine men regard them as "abodes of spirits." Other 
photographs show sacred dances, art, and architecture, and rituals associated 
with hunting and fishing. The contributions of second author, Robert F. Raffauf, 
give insight into the biodynamics of Banisteriopsis caapi, (Malpighiaceae) "vine 
of the soul" (from the Quichua name, ayahuasca) and other "plants in cosmic 



communication 



f t 



format 



accompanied by an extended commentary and a quotation from the writings or 
lectures of some of the anthropologists, explorers, ethnobotanists, and others 
who have worked in the Colombian Amazonia. Although some of the quotations 
date from the early 1900s or before, many are from studies of the last three 
decades, e.g., those of Michael J. Harner, Walter F. Lewis, Memory P.F. Lewis, 
Tommie E. Lockwood, Timothy Plowman, Melvin J. Shemluck, and Gerardo 
Reichel-Dolmatoff, to whom the book is dedicated. These singularly appropriate 
commentaries and quotations are, in and of themselves, lessons in ethnobotany. 
These texts and photographs are so well suited to one another, each comple- 
menting and elucidating the other, that the media seem destined to blend giving 
each of the facing pages an all but unified presentation— a manifestation of the 



a work of art and an ethnobotanical workshop. 



>/ 



Both the Foreward (by Ghillean T. Prance) and the Introduction present 

background material. Features not found in the companion book but 
welcome additions here are an Epilogue (by Michael J. Balick), a 



Quotations, giving complete bibliograph 



accompanying 
most of these 



probable that because of acculturation of the indigenous people, habitat destruc- 



com 



photographs could be taken today. How fortunate we are that Professors Schultes 
and Raffauf have shared these treasures with us and that Synergetic Press has 
offered them to us in book form at such a bargain. 



LITERATURE CITED 



SCHULTES, RICHARD EVANS. 1988. Where the Gods Reign: Plants and People of 

the Colombian Amazon. Synergetic Press, Inc., Oracle, AZ. 
VAN ASDALL, WILLARD. 1989. Book review of Where the Gods Reign: Plants and 

People of the Colombian Amazon, by Richard Evans Schultes. Journal of Ethno- 



biology 9:156-157. 



Willard Van Asdall, Past E< 
Journal of Ethnobiology 
4479 N. Summer Set Loop 
Tucson, AZ 85715 



234 



BOOK REVIEW Vol. 12, No. 2 



BOOK REVIEW 



M 



$26 



markable 



lm 



umes 



the new breed of naturalists who go there to conduct intensive, long-term field 
research rather than cursory collecting expeditions. Beehler has spent a total of 
almost five years in New Guinea, spread over eight visits beginning with his 1974 
study of bird of paradise courtship behavior, but increasingly focusing on eco- 



logical topics. 



many technical publications 



audiences, offered as "lay introduction to the island's natural history and a 
sort of catalog of the things naturalists do when out in the forest" (p. 12). The 



more 



numerous 



panying each chapter and a separate section of color plates. Interspersed among 
the field stories recounted are illuminating examples of how research problems 
can be redefined in the field. For example, while pursuing the question of why 
some bird of paradise species are monogamous and others polygynous, Beehler 
found he had to study their food resources as well, resulting in a realization that 



more 



dissemination 



lm 



Papuan environment" (p. 139). 

Beehler' s insightful reflections on his work and experiences are relevant to 
regions far beyond New Guinea, and he is especially effective at disabusing the 
reader of simplistic notions. Thus, despite the fact that his main ornithological 
interest is in some of the world's most spectacular fauna, he cautions that "[the] 
real jungle, most of the time, appears dull and dark. Much of each day the forest 
is silent and there is not a bird to be seen for love or money. Perhaps that is why 
for so many decades the rainforest has received little attention from the world 
at large. The rewards are hard- won" (p. 205). 

For the ethnobiologist reader, there is much to learn and enjoy in Beehler s 
book, but one glaring omission is continually evident. While he periodically refers 
to his local field assistants, and argues for increased training of "promising 
indigenous naturalists" (p. 243), nowhere in the book does Beehler indicate that 
he learned anything from them. "When food acquisition is concerned, the New 
Guineans leave nothing to chance-this is not sport, after all" (p. 216): all the 
more reason to suppose that their folk biological knowledge has remained a vast 
resource inexplicably untapped bv this otherwise meticulous and keen observer. 



Terence E. Hays 
Rhode Island College 
Providence, RI 02908 



/. Ethnobiol. 12(2):235-270 



Winter 19 




•If 



ABSTRACTS 

of presentations (contributed papers and poster sessions) 

at the Fifteenth Annual Conference 

of the Society of Ethnobiology 

National Museum of Natural History 

Smithsonian Institution 
Washington, D.C. 
25-27 March 1992 



Mary J. ADAIR, University of Kansas, Museum 

Hall, Lawrence, KS 

PREHISTORIC AGRICULTURE IN THE CENTR 

RESEARCH AND FUTURE DIRECTIONS 



To date, the earliest evidence for domesticates in the Central Plains is 
marshelder (Iva annua v ar macrocarpa) from Early Woodland deposits dated ca. 
500 B.C. Corn (Zea mays), squash (Cucurbita pepo), gourd (Lagenaria sicerana), 
sunflower (Helianthus annuus var macrocarpa), beans (Phaseolus vulgaris), and 
tobacco (Nicotiana sp.) appear in the following cultural periods, such that by 
A.D. 900, an agricultural economy is recognized. Initial research, which focused 
primarily on chronology, sought to identify specific cultigens and to document 
their distribution throughout the Central Plains so that a relative importance on 
agricultural crops could be established for various cultural periods. A refinement 
of this chronology and a better understanding of economic patterns are currently 
underway. Efforts are underway to direct date all reported early cultigens as well 
as those with questionable cultural context. Stable carbon and nitrogen isotope 
studies may evaluate the relative dietary importance of maize. An explanation 
of the variability within the maize remains themselves, also a current research 
focus, involves the analysis of curated collections, ethnographic collections, and 
modern Native American carden varieties. While these topics will continue as 



some time 



Way 



information may 



Natalie P. ALEXANDROVICH, Republic Byeloruss, Institut Istorii ANB, 



, .., Minsk 

ANIMALS AS THE TOTEMS 
POPULATION 



Amulets made from bones, horns, and ceramics 



archaeological materials. The most ancient amulets are little ducks made from 



ceramics 



with holes and grooves for'lacing, are known from many archaeological sites. 



236 



ABSTRACTS Vol. 12, No. 2 



It is important to emphasize that amulets of different totems, such as beaver (Castor 
fiber L.), bear (Ursus arctos L.), various birds, and carnivores are found. The most 
common totems are bears, beavers, martens (Martes martes L.), polar foxes (Alopex 
lagopus L.), wolves (Canis lupus L.), badgers (Meles meles L.), wild boars (Sus scrofa 
ferrus L.), and grass-snakes (Natrix natrix L.). The totems of domestic animals 
appear later in the archaeological record. Amulets made from the teeth of domestic 
pigs have been recovered from early pheodalic sites from southeast Byelorussia 
(Nisimkovichi). Animal totems occupied an essential position in the life of the 
slavic tribes of Byelorussia. As documented in archaeological materials, totemism 
began in the paleolithic and was restricted to middleaged individuals in cultic 



ceremonies. 



KAMEL 



Mexico 



^ • 



04460, Mexico 



FIREWOOD 



NAHUATL GROUP IN GUERRERO, MEXICO 



In this paper we assess the ethnofloristic resources of the basin of the Balsas 
Medio, Guerrero, which were used by the Nahuatl indigenous group. The Nahuatl 
are spead over the central section of the state of Guerrero in small villages of less 
than 3000 inhabitants each, including Xalitla, Ameyaltepec, S.J. Tetelcingo, S. 
Agustin Oapan, and S. Fco. Ozomatlan. Wood was used for firewood (domestic 
and ceramic firing), as well as for the manufacture of handicrafts. 

The following species have been identified as primary vegetation in Guer- 
rero: deciduous tropical forest (BTC) with dominance of burseras, BTC with 
dominance of columnar cactus known as noxtli (Neobuxbaumia mezcaletisis) , and BTC 
with the palm Brahea dulcis (zoyatl). Secondary vegetation is mainly composed 



estimate the production ootential of some 



sam 



w 



hich 



(domestic), cuajiotes 



spp.), which is used for handcrafts, and noxtli (N. mezcaletisis) that is used to make 



volume 



hectare. 



m 



Chevy Chase, MD) 



Miraflores. Lima. Peru (5480 Wisconsin 



STRATEGIES 



SOUTH AMERICA 



America 



and 



^ j_.^..,_j, uc U31 ui is seriously nnpcicis uic c*-«-"-~ 

public health of indigenous populations. Coca (Erythroxylon coca) cultivation and 
the polluting effects of its transformation into PBC and cocaine, along with the 
consequences of its marketing for the societies taking part, is the most obvious 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 237 



recent example. When the effects of cattle raising and temporary agriculture and 
the resulting alteration of the rainforest are considered, it is evident that the only 
real management alternative is reforestation with native species and native crops. 
Where such biotic communities generate naturally produced products, primary 
forests must be protected along with the established biotic variables that would 
favor their maintenance. This would be nothing new for the native societies liv- 
ing in the forest, who are being pushed to the margins. Unfortunately, their 
knowledge, which has developed through the centuries, is not taken into account. 
The Biological Expositive Unit at Huaca Pucllana and its archeozoological and 
archeobotanical laboratories are being created to record and rescue information 
regarding the management and knowledge of plants and animals used by 
prehispanic societies— their nutritional, medicinal, and industrial uses, and their 
symbolic iconographic use in the arts. This research is carried out in an ecological 
context that is essential to providing governments, financial institutions, and 
people an understanding of the perils of continuing with the present defores- 
tation rate. 



Museum of Anthropology, University of Michig 



MI 



FLOW 



CONCEPTS OF Zea mays 

Existing anthropological analyses of Hopi concepts of corn have been hindered 
by the narrowness of our own interpretive constructs, which emphasize the 
materialist concerns underlying the themes of nourishment and reproduction. 
As a consequence, we have a poor ethnological comprehension of the central role 
that corn concepts play in conditioning how the Hopi understand and occupy 



Washes 



metaphors can be made more com 



damental im 



Alejandro de AVILA B., Instituto Tecnologico de Oaxaca, Apdo. Postal 1378, 



:. C.P., 68000, Mexico 

CONTEMPORARY MIXTEC RITUAL: f uncus, Nicotiana 



and Solandra 






The Mixtec comm 



westernmost Oaxaca State, in southern Mexico 



m about 1000-3000 m 



The local flora includes over 1000 



plants. In this paper I describe the ritual use of four species, two of which are 



home 



(Juncus) and forests (Solandra) above the village. Bundles of rush () 



i rain-petition ceremony, celebrated on certain muu 
Marcos). Locally grown tobacco is associated with m 



harm 



238 



ABSTRACTS Vol. 12, No. 2 



for divination, particularly after the loss or theft of property. Seeds of a fifth, 
unidentified species, possibly Turbina, were used similarly. The latter plant does 
not grow locally, but the seeds were brought in from the Pacific lowlands. I review 
pre-Columbian codices and sixteenth century Colonial sources to document the 
ancient use of rushes and Datura in Mixtec culture. I also trace the etymology 
of the Mixtec names of the five species in Coicayan and compare them with terms 
used in other Mixtec languages to indicate their salience in Mixtec ethnobotany. 
Finally, I compare the rush bundles with the similar bunches of pine and copal 
resin employed by the Mixe and Chontal peoples of Oaxaca, and I relate the use 
of Nicotiana, Datura, and Solandra to the knowledge of hallucinogenic Solanaceae 
elsewhere in Mesoamerica. 

Bradley C. BENNETT, The New York Botanical Garden, Institute of Economic 
Botany, Bronx, NY 

PLANTS AND PEOPLE OF ECUADOR'S AMAZONIAN RAINFORESTS: 
LESSONS AND NEEDS FOR SUSTAINABLE DEVELOPMENT 

Ethnobotany has helped develop many important Amazonian products 
including rubber, auinine. chocolate, and curare. Native plants remain vital 



many 



medicines 



Quichua use 90.9% of the species found in 

Amazonian Ecuador's native neonlp use nea 



derestimate 



lm 



Ethnobotany not only identifies plant resources but also the indigenous 



them 



for century-old techniques. Indigenous people plant, protect, or collect plant 



resources. Combinations 



ems of forest use. Some of these systems 



alternatives to destructive practices currently used. 

Much needs to be done before we can realize ethnobotany's full potential 
for protecting the plants and people of the Amazon. First, we must continue to 
identify the plant resources. Both rainforests and cultures are disappearing rapidly. 
The coexistence of intact cultures and intact forest is exceedingly rare. Second, 
we must evaluate the economic potential of forests and promote the marketing 
of nonwood products. Third, we should promote new uses for traditional pro- 
ducts. Carludovica palmata R. & P. is the source of the famous "Panama" hat. Yet, 
it is as important in traditonal societies for food, shelter, and basketry. Many 
development efforts in Amazonian have been cultural and ecological disasters. 
There are several wavs wp ran av™H fho ration* n ( ~-™,;«,,o «v™o/-tc Promotion 



multiple species is one way. Much 
a single resource. Multipl 



kji me past rauure is uuc iu u» - — 

, ecies are especially desirable. A secon 
promote products at the local level. As these develop 
we can expand to regional, national, and international markets. Grassroot sup- 
port within communities will help assure success and promote benefits to com- 
munity members. 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 239 



Antonio BIANCHI, Yia Sommariva 5, Verona, 37131, Italy and 

Samorini GIORGIA, Natural History Museum of Robereto 

PLANT USE IN THE AYAHUASCA COMPLEX IN NORTH WEST AMAZONIA 



Contemporary 



many 



vegetal hallucinogens, others could influence the course of the Ayahuasca intox- 



ication, 



We 



from 



armaco 



of this kind of work is to establish, when possible, the level of study regarding 
any plant and to give prominence to the importance of further fieldwork. 

Zlatozar BOEV, Bulgarian Academy of Sciences, National Museum of Natural 
History, Boulevard Russki, Sofia 1000, Bulgaria 

RAPTORS AND OWLS (FALCONIFORMES ET STRIGIFORMES - AVES) 
FROM ARCHAEOLOGICAL EXCAVATIONS IN BULGARIA 

All archaeozoological data of diurnal and nocturnal raptors (vultures, eagles, 
falcons, hawks, and owls) from Bulgaria are presented. A total of 4685 bones and 
bone fragments of birds were collected from 29 archaeological sites in the coun- 
try. In 16 of them, 77 bone remains (1.64%) of raptor birds are established. Sites 
cover a very long period— from the Upper Paleolithic (ca. 31,900 years B.P.) to 



Medieval 



21 species, 17 raptors and 4 owls. The most numerous are the griffon vulture 



The 



are 



Lammergeie 



Among 



falcon, hen harrier, honney buzzard, hobby, kestrel, red-footed falcon, little owl, 



Lammergeie 



Bulgaria. 



All these wild 
/. Some of them hav 
companions of man 



endangered and rare 



m 



Zuni, NM 87327 

PRESERVING TRADITIONAL CROPS IN RURAL 

COMMUNITIES: AN EXAMPLE FROM THE PUEl 



Mexico 



in me lasi one nuiiureu ycdis mc i uv^ ". 

has been confronted with a reduction of traditional lands, the mtroduct.on of a 
cash economy, and the destruction of valuable farmland. Crops that once sus- 
tained the Zuni people are now threatened with extinction. To prevent the loss 



240 



ABSTRACTS Vol. 12, No. 2 



of these resources, the Traditional Zuni Crops Project was undertaken in the Fall 



of 1991. The goals of this project were to document crop diversity on the Zuni 
preservation and to collection donations of traditional Zuni crops for preserva- 
tion in a tribal seed bank. Fifty households were interviewed concerning crops 
grown from year to year in their fields and gardens, as well as the source of seed 
stock for each crop variety. The survey found that those traditional crops that 
are more common in the Zuni community are those that are closely associated 
with the Zuni religious activities and ceremonial calendar. Attitudes towards 
sharing seed through a tribal seed bank were also documented, revealing the role 
of seed in many of the winter ceremonies. These data emphasize that traditional 
crops are perceived by the community as an important cultural resource, integral 
to Zuni cultural identity. Insights such as these are an important preliminary step 
in designing a local seed bank program that will be able to serve the community 
in ways that the people recognize as being appropriate and useful. 

Cecil H. BROWN, Department of Anthropology, Northern Illinois University, 

Dekalb IL 60115 

LEXICAL ACCULTURATION IN NATIVE AMERICAN LANGUAGES 

This study identifies and explains cross-language patterns in ways Native 
American languages have named new plants and animals and other objects and 



Pt 



items 



(e.g., rice, sugar, sheep, and horse) in 



number of American Indian langu; 
from Spanish into Latin American 



terms manufactured from native vocabulary (e.g., literally, "little maggots" for 



number of North American 



for 



example, that languages that are heavy adopters of loans for these items have 
histories involving significant bilingualism of speakers, while languages that show 



terms for such items 



some items 



items 



items 



may be related to the possibility 



donors in the past to a greater extent than have words for items such as hen and 
rooster. Elucidation and explanation of these and other findings will contribute 



areas 



language universals, language and culture change, and social histories of indi 
vidual Native American groups. 



Carmen 

London, CT 06320 



Connecticut College, Box 5427, 270 Mohega 



FUZZY FIELDS AND CATCHALL CATEGORIES: THE CENTRALITY OF 
INDETERMINANCY IN TO MAKI SWIDDENING PRACTICES 

Ethnoecological research methods lone have been deemed 



producing static classifications 



dynamic and pro 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 241 



cessual aspects of human-environment interactions. This paper asks: Is this 
intrinsic to the methodology or due to the ethnographer's primary emphasis on 
identifying attributes that clearly distinguish discrete entities and categories? In 
fact, has the "native point of view," a view often more inclusive in what it con- 
siders and encodes, been thoroughly explored in relation to classifications? 
Based on fieldwork with the To Maki Toraja, upland forest cultivators living on 
the island of Sulawesi (Indonesia), this paper looks first to To Maki classifica- 
tions of garden types. While contrasts between garden types are readily made, 
in practice the To Maki are more attentive to the fluidity of boundaries between 
types and catchall categories— the points at which classifications get "fuzzy." The 
paper then focuses on the pa 'lak, the most variable garden type identified by the 
To Maki. Although the pa'lak is indeterminate in its formation, cropping pattern, 
and life-span, as an element in the swiddening regimen this "fuzzy" field is 
central in allowing the To Maki to position themselves to respond quickly to uncer- 
tain future states. In this the pa'lak seems key in the overall operation of the To 
Maki subsistence economy. By looking more carefully at the penumbra of mean- 
ings associated with classifications, along with the points of transition between 
categories rather than just points of contrast, ethnoecological methods can be 
salient in identifying major environmental problems faced by subsistence 
cultivators and serve as important guides in deciphering subsistence histories. 



Robert A. BYE, Jr., Jardin Botanico, Instituto de Biologia, Umversidad Nac 
Autonoma de Mexico, A.P. 70-614, Ciudad Universit., 04510, DF, Mexico 
Peter KEVAN, Department of Environmental Biology, University of Gue 
ETHNOBIOLOGY OF MADRONE BUTTERFLY (Eucheira socialis 
[LEPIDOPTERA: PIERIDAE]) OF MEXICO 



The madrone 
primitive morpho 



The 



madrone tree of the Sierra Madre 



Neovolcanic Mountains of Mexico 
two- walled bae composed of doul 



from September to May 



larvae 



manufacture 



from the silk-like bag. Contemporary 

Mexico 



of Chihuahua) is consumed by the older people but rarely by the young. The 



May 



calorie source for these traditional subsistence agriculturalists during the end of 



dry 



Tarahumara 



mvtholoev of mo 



in Mesoamerican traditions based upon insect metamorphosis. The 



242 



ABSTRACTS Vol. 12, No. 2 



imprudent destruction of madro 
while neglect of the management 



madrone-z'iw'fc 
il information 



CANTLEY and Leslie E. RAYMER 

• Mountain, GA 30083 



WETLAND HAB1 
ENVIRONMENT 



Archaeological investigations undertaken in the interior uplands of the 
southeastern Piedmont and Appalachian regions find that the Archaic Period sites 
tend to cluster in and around swamp environments. Traditionally, archaeologists 
have interpreted this pattern solely on the basis of the subsistence component 
of these cultural systems. Our investigations of sites in these localities suggest 
that other variables interact with the subsistence component in determining 
swamp margin prehistoric settlement patterns. This paper will explore both the 
economic and social components of hunter-gathered systems and their interac- 
tion with the ecological parameters of wetland environments. We hypothesize 
that Archaic Period hunter-gatherers established their settlements in and around 
upland swamps in order to fulfill basic subsistence needs and to aid women in 
their dual roles as child-care providers and plant food collectors. 

A brief examination of 14 modern hunter-gatherer groups suggests that 
women in these societies make significant subsistence contributions and are 
probably nursing children for most of their adult lives. Given the high energetic 
costs of continuous lactation and the importance of women's subsistence con- 
tributions, it would have been advantageous to Archaic groups to locate their 
settlements where women could effectively and efficiently perform both food 
gathering and child care tasks. A review of the botanical and ethnobotanical 
literature for the southeast suggests that swamp habitats represent such localities. 
Swamp environments in the study area are characterized by high species diver- 
sity and productivity and are rich in edible and medicinal plants. Swamp margin 
camps would have allowed women easy access to a wide variety of economically 
important woody and herbaceous plants. This easy access would minimize 
women's energy expenditures on plant gathering forays, and would enable them 
to use this "saved" energy in the care and nurturing of their children. Our 
research shows that hunter-gatherer mobility-settlement strategies may vary 
along several ecological and social dimensions. As such, our investigations may 
provide important new information for addressing not only which cultural com- 
ponents are present on swamp edge sites but the organizational and structural 
characteristics of these sites as well. 



Wesley COWAN, Cincinnati Museum of Natural History, 1720 Gilbert Ave. 

icinnati OH 4S?n? and Rmro n ciVifrnj r* ■. * ~c a ^tu^nnlnav. NMNH 



MRC112, Smithsonian Institution, Washingt 



Bruce D. SMITH, Department of Anthropology 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 243 



WILD 



EASTERN NORTH AMERICA 



Free-living Cucurbita pepo gourds have been documented in disparate areas 
in the temperate eastern woodlands since the first few decades of the nineteenth 
century. Until recently, however, little attention has been afforded these popula- 
tions, and specifically, their potential role in the evolution of domesticated forms 
of pepo squashes has been minimized. The paper focuses on the botanical history 
of free-living gourds, examines the availability of cultivar ornamental gourds to 
American gardeners, and summarizes habitat information for contemporary 
populations in several areas in the east. Important morphological measurements 
from both contemporary free-living and cultivar pepo gourd populations as well 
as archaeological specimens are compared. These combined data suggest that 
free-living Cucurbita pepo gourds are ancient members of an eastern flora, and 
probably played an important role in the evolution of domesticated squashes. 



D.S. DECKER-WALTERS, T. WALTERS, Fairchild Tropical Garden, 11935 Old 
Cutler Rd., Miami, FL 33156, C Wesley COWAN, Cincinnati Museum of Natural 
History, 1720 Gilbert Ave., Cincinnati, OH 45202, and Bruce D. SMITH, Depart- 
ment of Anthropology, NMNH, MRC112, Smithsonian Institution, Washington, 
D.C. 20560 

CHARACTERIZATION AND INTERPRETATION OF ALLOZYME PROFILES 
IN WILD POPULATIONS OF Cucurbita pepo 



from 20 wild populations of C. pepo ssp. ovifi 



Missouri 



Comparison of these data to simil 



in Texas (C pepo ssp. ovifera var. texana) and Tamaulipas, Mexico (C. pepo ssp. 
fraterna), Mexican landraces (C. pepo ssp. pepo), and cultivars representing both 
major genetic lineages (C. pepo spp. ovifera var. ovifera and C pepo ssp. pepo) 
revealed a distinct allozyme profile including the characteristic allele Idh-2m 
for the non-Texas U.S. populations. Although about 50% of these populations 
exhibited signs of limited, and, in most cases, recent introgression from cultivars, 
only one population, which was represented by a single fruit from Kentucky, 
deviated sufficiently from the typical allozyme profile to be considered a possible 
escape. Both populations from Louisiana possessed Idh-3o, an allele otherwise 
restricted to the Texas population. All U.S. populations were distinguished from 
Mexican and domesticated C. pepo by having relative! 



Mexico 



me 



me 



over thousands of years in at least three disjunct ecogeographic habitats (north- 
eastern Mexico, Texas, and the midwestern United States) within the range of 
C. pepo. Furthermore, cultivars were selected independently at least twice, once 
from populations in Mexico and once from populations in the United States. 



244 



ABSTRACTS Vol. 12, No. 2 



Lin DUNBAR, University of North Carolina, 617B Hibbard Dr., Chapel Hill, 
NC 27514 

THE LIVE OAK: SYMBOL OF THE SOUTH CAROLINA LOWCOUNTRY 



Carolina lowcountry. Though the meanings 



multivalent symbol in the imagery 



communicate, and reinforce 



i. Within the lowcountry's natural, cultural, and societal history, the live oak 
dramatic public symbol that serves a leading role in the region's presentation 
self. While many individuals acquiesce to the tree's public image, others 
e personalized and transformed it into a statement of counter- hegemony. This 
>er explores the image of the live oak, emphasizing its counter-hegemonic 



meanings 



deFRANCE, Department of Anthropology, Florida Museum 



History, University of Florida, Gainesville, FL 32611 



IMPERIALISM 



MOQUEGUA 



SETTLEMENT 



accompanied by dynamic 



Peruvian 



farming and livestock rearing were im 



settlement 



the only American center of prehistoric large mammal domestication, 
introductions of Old World domesticated animals for both subsistei 
industrial purposes led to the emergence of unique patterns of colonia 
use. The economic and environmental repercussions of the establishment 



erpr 



many 



prolifi 



impact of Spanish-introduced fauna in the settlement 



am 



from the Moquegua Valley of far southern Peru. The 
. ammal use from the sixteenth through the nineteenth 

These 



s 



c . , . -■©«»• ^«"c udict are usea to assess me cucu ~* 

bpanish-introduced fauna on the peoples and habitats of the central Andean 

rpmnn 



region 



Mulford 



EVERETT, Department of Forestry and Resource Management 



PALM 



SRI 



HIGHLAND 



Caryota urens, known as the kitul palm 



most common trees in the multi-species 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 24^ 



homegardens of the highlands. Along with the palmyra in arid areas and coconut 
in the coastal zone, kitul is one of three palms traditionally tapped for nectar from 
which sweet syrup, sugar, and various alcoholic beverages are prepared. The 
syrup and sugar have a special richness akin to maple products known in the 
United States, and are highly valued for culinary and medicinal purposes in Sri 

Lanka. 

Tapping palms is the domain of an occupational caste of "toddy tappers" 
who traditionally divide the yield with palm owners. This paper focuses on the 
tappers' knowledge and management of kitul palm, its products, and tapping 
and processing activities. The ecological and economic importance of developing 
markets for kitul products in relation to forest and landscape conservation in 
highland Sri Lanka is discussed. This paper is based upon the author's interviews 
with tappers, study of kitul yields, and participant observation while living in 
villages during dissertation field research on home gardens in highland Sri Lanka 
in 1989-91. 



Catherine S. FOWLER, Department of Anthropology, University of Nevada, 

Reno, NV 89557-0006 

NORTHERN PAIUTES AND THE BIRDS OF STILLWATER MARSH, 



NEVADA 



Marsh 



in western Nevada for at least 1000 years. During that time, they have observed 



mierations of numerous 



from 



learned to duplicate their calls, and so on. They have also witnessed the destruc- 



lm 



som 



.__ r , hat waterfowl played in their subsistence cycle 

and in their broader orientation to their lands and world. Ethnographic data are 
drawn from the extensive knowledge of Wuzzie George, who witnessed the 
transition of her people from native foragers to forced farmers in the latter part 
of the nineteenth and early twentieth centuries. 

Jose GONZALEZ RODRIGO, Universidad Autonoma Metropolitana, Iztapalapa, 
D.F., 09340, Mexico, and Regina Leal GUEMEZ, Unidad Iztapalapa, Iztapalapa, 

D.F., 09340, Mexico 

DEVELOPMENT OF AN EXPERT SYSTEM FOR THE STUDY OF 

THE MANAGEMENT OF NATURAL RESOURCES IN INDIGENOUS 

COMMUNITIES 

There is in ethnobiology a marked interest in the study of the management 
of natural resources in peasant indigenous communities from an holistic outlook. 
In this case, it is imoortant to emphasize the role that peasant economies play 

rn societies. Therefore, it is essential to have 



development 



me 



246 



ABSTRACTS Vol. 12, No. 2 



munities. The objective of this paper is to introduce a model for the m 



community performe 



ems. With this mo 



economics 



udied community. This model is built from invest 
Monte, Texcoco, located in the Vallev of Mexico 



M. Johnson GOTTESFELD, Department 
a, Edmonton, Alberta, T6G OR2. Canad, 



NORTHWESTERN BRITISH COLUMBIA 



MANAGEMENT 



The Giksan and Wet'suwet'en peoples of northwest British Columb 



River 



drainage 



transitional between the northwest coast and the boreal interior. The land is 

m • - 



mountainous 



Wet'suwet'en for veeetation mani 



Berry patch burning was the most important traditonal vegetation manipu- 
lation. Montane black huckleberry (Vaccinium membranaceum) and lowbush 
blueberry (Vaccinium caespitosum) patches were burned in the fall to stimulate 
growth of new stems and production of berries, while preventing invasion by 
other shrub species and young conifers. Low elevation berry patches might also 
be burned in the spring. Soapberries (Shepherdia canadensis) are also reported to 
have been managed by burning at times. Burning of berry patches is reported 
to have been done by groups of women, or by groups of men on the way to 
hunting mountain goats in areas above the berry patches. Berry patches were 
owned and were located within house group territories. Only the owners had 
the right to burn their area or to arrange for it to be burned. 

Berry patch burn intensity was controlled by utilizing appropriate seasonal 
and diurnal timing, knowledge of weather conditions, and fuel loading. 
Maintenance of berry patches b V burnine was dismntimioH in th* 10*)* and 1940s 



mg 



Columbia 



continues to the present. It occurs in aspen, pine, or grass-dominated sera 
communities, or cottonwood floodplain forest, and is intended to control brusl 
and encourage growth of grass or garden vegetables. 

^Tu ^ N J HAM ' Jessica BOOHER, Bill WEYLMAN, and Joel BARNES 

cllfr * c x , Studi6S ' Marine Mammal Biology and Conservation, Baja, 
IMN^TFir IV^KT C °^i Grantham: m ° v « hfll Road ' Baltimore, MD 21210) 

bEE^VZSIS? FISH species off the shores ° f espiritu 

IelY^Vf IMPAIR AN ° L ° S ISLOTES ISLAN °S AND THEIR 
faHfnl T* ?* TANCE T ° ™ E CALIFORNIA SEA LION (Zalophus 
caltformanus) AND THE FISHERMEN OF LA PAZ 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 247 



Fishing techniques and catch composition of artisenal fishermen were estab- 
lished from Espiritu Santo, La Partida, and Los Islotes Islands of Baja California 
Sur, Mexico, between the months of September and November 1991. Twenty- 
two samples from drift nets, seine gillnets, and hook and lines yielded 35 fish 
species. Relative number, weight, and catch frequency per species were calculated 
to identify those most important to fishermen. Barred pargo (Hoplopargus guen- 
theri), broomtail grouper (Mycteroperca xenarcha), and bronze-striped sea chub 
(typhosus analogus) were found to be most important, followed by greybar grunt 
(Haemuoln sexfasciatum), pacific goldeneyed tilefish (Caulolatilus affinis), and red 
snapper (Lutjanus peru). Four of the 35 species caught by fishermen matched prey 
species of the California sea lions (Zalophus californianus) inhabiting the Los Islotes 
rookery. Of these species, only Caulolatilus affinis had a high importance to 
fishermen. This species is a minor part of the sea lion diet, suggesting minimal 
human/sea lion competition. However, seasonal changes in fishermen's catch and 
sea lions' diet may create competition at other times of the year. This study 
provides baseline data for future management plans to sustain artisenal fishing 
and a healthy sea lion population in the Bay of La Paz. 






Robert J. HARD, Division of Behavioral and 
Texas at San Antonio, San Antonio, TX 78249 



MOBILITY 



ADOPTION 



Recent advances in the cultural ecology of hunters and gatherers suggest that 
there are particular ecological characteristics which encourage trends toward 
reduced residential mobility and increased use of cultigens. Comparisons of a 
number of prehistoric trajectories indicate the potential for identifying common 
ecological factors which influence the adoption of agriculture and reduced 
mobility. 

David G. HILL, Center for Energy and Environment, University of Pennsylvania, 

Philadelphia, PA 

ENERGY USE AND MANAGEMENT IN A NEPALI HILL VILLAGE 

Energy resources utilization is a major interface between humans and the 



environment 



middle 



in a village in rne miauie run icgiuns ui iNcpa*. »»,«.»*« 

based first upon an understanding of the structure and genesis of local resource 
management systems, and then upon a practical inquiry into how this knowledge 
translates into action. Ethnobiologists have a role to play in outlining for various 
agents the dynamics of the interface between humans and biomass resources. 
I set out with the objective of understanding local perceptions and knowledge 



em. and evaluating alternatives. While 



predominant 



more 



248 



ABSTRACTS Vol. 12, No. 2 



it did on the forest or other biomass resources. Certainly indigenous awareness 
and knowledge of the factors that shape the biomass use system are extensive, 
and rarely well understood by outsiders, but I took this observation as an indica- 
tion of local priorities and realities. A respected informant suggested at one point 
that rather than weighing and measuring of wood loads and other biomass 
flows I should devote more investigative effort towards current and historical 

local politics. 

Villagers in the study area are highly aware of resource use and management 
issues and conflict. Forest tenure is dominated by patterns established many 
decades ago, that do not reflect current forestry law. Vested interests, practical 
conservatism, and limited resources devoted to policy implementation and 
monitoring all play a part in forming this gap, which is probably common, 
between the de facto and de jure situations. 

While more than 95% of the total energy supply in the study area comes from 
biomass resources, an investment in a micro hydro-electric plant, that would 
replace imported fossil fuels used for lighting and agricultural processing, is 
currently the most feasible and attractive energy management option. This find- 
ing is based on the relative opportunity costs of capital and energy resources in 
the village, as well as consideration of the social and political factors that shape 
the current biomass use system. The research does not suggest that opportunities 
to improve the productivity, equity, or sustainability of biomass resource use and 
management do not exist, or are not needed. It is critical to acknowledge, however, 
that even when biomass plays a predominant role in a particular energy system, 
investments in the resource must be weighed against other options. 

Many countries are deeply involved with policy issues concerning biomass 
energy resources. Governments gave the challenge of increasing their role of 
facilitating and empowering development, rather than serving primarily as sources 
of first and last recall or as conduits for foreign aid. Forest policy makers must 
evaluate the relative costs and benefits of common versus private property 
management, and the nature of interactions between local systems and external 
agents. This research uses an "ethnoenergetic" approach to further the under- 
standing of these and related topics. 



HUANG, Xiecae (HUANG, Se-zei), Guangxi Institute for Drug Control, Guangxi 
China 

A PRELIMINARY STUDY OF Tiancha (SWEET-TEA) OF GUANGXI 
PROVINCE, CHINA 



commercial name 



ethnomedicine 



main material to make drinks 



medicine. It has been pur 



commercial 



sample of commercial drug and specimens. It has been found that the botanical 
origins of tiancha (Sweet-tea) in Guangxi are four species. These are: Rubus 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 249 



suavissimus S. Lee (Rosaceae), commercial name yaoshan-Hancha; Lithocarpus 
polystachyus (Wall.) Rehd. (Fagaceae), commercial name baise-tiancha; Engelhardia 
roxburghiana Wall. (Juglandaceae), commercial name guiping-tiancha; and Mycetia 
sinensis (Hemsl.) Craib (Rubiaceae), commercial name longzhou-tiancha. 



Neil C. HUGHES, Department of Anthropology, NMNH, MRC 112, Smithsonian 
Institution, Washington, D.C. 20560 

FAUNAL EVIDENCE OF THE EFFECT OF THE SPANISH CONQUEST ON 
INDIGENOUS PROCUREMENT ACTIVITY AT MANAGUA, NICARAGUA 

A small faunal collection excavated by salvage archaeology in Managua, 
Nicaragua, was divided into prehistoric and historic assemblages based on the 
introduction of European domesticated species. Faunal data provide evidence that 
the preconquest indigenous population practiced a broad-based procurement 
strategy which included 23 taxa and 32 minimum number of individuals (MNI), 
with biomass representing four vertebrate classes (mammals, fish, reptiles, 
amphibians). Also urn burial and burned human bones in a midden context are 



m 



assem 



MNI 



(mammals) of European domesticates 
:tion, the limited areal coverage it re 



P 



procurement and ritual activities. In Nicaragua, which is largely unknown 



lm 



framework 



confirmed 






establishing of allometric values by the Zooarchaeological Laboratory of the Florida 
Museum of Natural History has provided such a framework. The resulting paper, 
therefore, represents a first systematic assessment of faunal remains in Nicaragua 
and proposes that changes in faunal resource exploitation strategies point to a 
radical change in the structure of procurement activity resulting from the Spanish 
conquest. This hypothesis is supported by ethnohistorical accounts of organized 



human 



balism 



WA 



Department of Anthropology, University of Washingt 



SPECIMENS 



STIMULUS MATERIAL 



importance of collecting voucher specimens 



Ail*- 11HL/W11U11V.C VJ. \.wnvvnii^j ,ww»-w-.~- —j 'II £\ 

is well recognized. However, collecting zoological vouchers-especially ot large 



vertebrates— may 



nimal 



stimulus m 



250 



ABSTRACTS Vol. 12, No. 2 



Sound recordings can be at least as reliable for species documentation as 
photographs, study skins, or skeletal specimens, and such recordings are easily 
copied and edited for use in naming tasks with consultants at a later time. Basic 
equipment and procedures involved in making and using such recordings are 
also described. 



Timothy JOHNS, Gaetan M. FAUBERT, Vivian FRANKLIN, School of Dietetics 
and Human Nutrition and Institute of Parasitology, McGill University, McDonald 

Campus, St. Anne de Bellevue, Quebec, H9X ICO, Canada, and R.L.A. 
MAHUNNAH, Institute of Traditional Medicine, Dar es Salaam 
ANTI-GIARDIAL ACTIVITY OF LUO GASTROINTESTINAL REMEDIES 



ailments 



shrimp 



bioassays of 30 species were conducted in the field with the purpo: 
ing a means of predicting anti-giardial activity. Crude methanolic 



sampl 



Samples active below 500 ppm include Albizia coriaria, Comoiphora afr 



Microglossa pyrifolia, Ozoroa micronata, Rhus natalensis, Solatium nigrum, Sonchus 
schweinfurthii, Vernonia brachy calyx , and Ximenia caffra. Several of the plants that 
show activity against Giardia contain saponins or steroidal alkaloids. Crude 
extracts and purified fractions of these olants were tested for antimolluscicidal 



anism 



Elaine JOYAL, Department of Botany, Arizona State University, Tempe, AZ 85281 
ETHNOECOLOGY OF Sabal uresana (ARECACEAE) IN SONORA, MEXICO 



most economically im 
:h western Mexico. Its i 



Mexico 



bution, natural history, use, and "management" are poorly known, however. 
Preliminary data on the use and management of S. uresana, both my observa- 



tions ana tnose ot the indigenous and mestizo communities that use it, are 
presented here. 

Different size-class palms within the same population may have distinct 
names and uses. People travel up to a day to obtain cojoyos, newly-unfolding 
leaves, which are valued for weaving mats, baskets, and hats. The summer 
monsoon season is the major harvest time for cojoyos. Palmas, fully-expanded 
leaves, are used for thatching and for broom manufacture. They are harvested 
only when they are readily accessible, during the fall or spring dry season. In 
addition, all palm leaves are harvested during the full moon to insure their lon- 
gevity. Only dead palms are cut for logs. Burning of palm stands is common. 
My continuing research is quantifying palm populations and investigating local 
management of S. uresana for possible in-situ conservation. The potential impact 
of an increasing demand forS. uresana products bv tourists needs to be assessed. 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 251 



Jay Bahadur Singh KARKI, Social Science Division, International Rice Research 
Institute, Los Banos, Manila, Philippines, and Madhav B. KARKI, Institute of 
Forestry, Pokhara Campus, Pokhara, Nepal 

POTENTIALS OF HOME GARDENS IN IMPROVING THE SOCIO- 
ECONOMIC AND ECOLOGICAL CONDITIONS IN THE KATHMANDU 
VALLEY OF NEPAL 



In this study the floristic composition of home gardens and the utilization 
of these species in relation to their cultural origin, ecology, and socioeconomic 
factors were analyzed. The study found that home gardens play an important 
role in garden economics and sustainability of farming systems. Two types of 
home gardens were identified— permanent and temporary. The subsystems were 
subsistence home gardens, and commercial home gardens. Mixed home gardens 
(permanent), opportunity home gardens, and shared home gardens were also 
identified. Permanent home gardens are strongly and traditionally linked with 
their evolution, and there is strong desire to sustain them, whereas temporary 
home gardens can be left fallow if no suitable tiller is found. These home gardens 
have a high floristic richness, providing families with numerous products to satisfy 
various needs: of 230 species reported; 70% were used as nourishment, 20% as 
fodder and fuel wood, and 10% had secondary usages. However, the species with 
highest densities and frequencies were the food plants. This paper will also 
describe the total litter fall from the tree canopy and its contribution to soil 
fertility. The exchange of plants and knowledge of plants by the families in the 
communities have made home gardens more floristically homogeneous. Home 
gardens are also a place of agricultural experimentation in which all the family 
participates. 



J 



mos, Manila, Philippines 

FARMING SYSTEMS AND SOCIOECONOMIC 



SYSTEM IN THE MID 



NEPAL 



This study analyzed farming systems and socioeconomic aspects of an 
indigenous sustainable management system in the mid hills of Nepal, and focused 
on land-use patterns and people's perception of, and attitutes toward, the 
systems. Primary data were obtained through interview and observation. The 
traditionally developed farming systems that were identified were home garden 
only, home garden and paddy, home garden and upland, and home garden, 
paddy, and upland. Analysis of farming systems focused on the dominant crop- 
ping pattern-single component, double component, and multiple cropping 
systems. The predominant land use systems associated with farming m the area 
included three traditional, locally developed agroforestry systems-agnsilviculture, 
silvipastoral, and agrisilvipastoral. Subsystems were: multipurpose trees on 
farmland, shelter for plantation crops, wind breaks and homestead garden, 



252 



ABSTRACTS Vol. 12, No. 2 



system. Farmers 



grazing system 



ems, althoueh several expressed openness to some 



significantly related to socioeconomic characteristics (i.e., education, income, and 
farm structure). Farmers' attitudes were related to education, occupation, income, 
household size, and farming system. Age, sex, caste, religion, farm size, and 
family work force had no bearing on perception or attitude. Recommendations 
based on the study were grounded on using an approach characterized by refine- 
ment or improvement of the existing land use system rather than radical transfor- 
mations from the outside. 



Margarit 



COLUMBIAN EXCHANGE OF MEDICINE 



Columbus's travels to the New World occurred at the time 



Columbus 



medicinal 



many plants and many 
medicines of spicery " 



>/ Marco Polo and was going to look for the plants he 
st report Columbus writPQ " what causes me the 



-«™ ^ ~ ~^ v ^v. xxi no moi ic^uii ^AJIUIIIUUS Writes . . . Wliai LdUSta uw **"" 

greatest grief in the world, when I see a thousand sorts of trees that each have 
their own kind of fruit, ... and a thousand sorts of plants, the same '" n " 



in 



much 
" "(M 



Columbus 



^v.vx lv * v^agc, ivi. i^iegu /\ivarez i^naiita waa ap^vui^ — 

official physician to the fleet. Chanca described native healing and natural history 



ethnography of the New World (Ybarra 1894, 



made 



same 



with in Spain ..." In New Spain, Cortex continued the medical ethnographical 
tradition. From letters we learn of his respect for Aztec herbs and also of his urgent 
request for bringing plants from Spain. The Columbian exchange of medicine 
has begun. By the 1540s, the Franciscan friar Bernard de Sahagun was collecting 



information on Mexic 



World" was sent 



new 



u ins team until iD/b. He described some 3076 plants, many of which were 
i medicines. His descriptions emphasized morphology of the roots and 
making it difficult to provide positive botanical concordances. Monardes 



Herbal (orig. 1633). 



7) on the medicinal plants that he had i 
appeared in Johnson's revision of John 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 253 



missioniz 



natural history followed that included each area's healing plants. And so con- 
tinued a orocess of medical acculturation wherein Old World and New World 



This 



arm 



KREMENETSKI 



Staromometny Lane 29, Moscow 



ENVIRONMENTS IN THE LOWER 



REGION 



A pollen analytical study of Kardashinski peat bog, situated near the mouth 

_ in i mm i • . * l -i A 1 « * 



Dniepr (46 degrees 31 'N, 32 degr 
made it possible to reconstruct the history 



8000 years. Up to 4200 BP the flood-plain and sandy terraces of the Dniepr were 



com 



birch, and especially pine. 
Later, forest areas diminished due to climate changes and human impact. In the 
lower part of the Dniepr Valley the largest forest and steppe areas extended 
northward to the Black Sea shoreline. In the vicinity of the Kardashinski bog, 
many settlements of bronze-smelters were discovered, dating to 3500-2900 BP. 
The location of these Bronze age settlements is well understood from the palaeo- 
botanical viewpoint. At the beginning of the Iron Age, when Scythian tribes and 
Greek colonies appeared northward to the Black Sea, the intensity of land- use 
seriously grew. In the middle of the fifth century BC "the father of history," 
Herodotus, visited the Greek city Olbia on the north shore of the Black Sea 
(46 31 'N, 31 40'E). In his description of the Black Sea region Herodotus mentioned 

a _ 4 w WW * 1 "1 il £ 1 \ mm* L. -«* mmm* W J V f I *^fc df~\ ^\. T A ' ■ L. 



mentioned 



geobotanical and soil data allow the placement of the Hylaea forest on the left 



Pi 



the ideas of both historians and botanists. In settlements of the Scythian period 
charcoal oiPinus, Betula, Quercus, and Corylus were found. In rural settlements 
of Greek period near the city of Olbia (400 BC-AD 400) charcoals of Ulmus, 
Fraxinus, Quercus, Tilia, Pinus, Populus, and Alnus were found. This correlates very 
well with the composition of pollen spectra of the diagram of the Kardashinski 

bog for the same period. , 

The existence of forests in the river valleys is also demonstrated by arcnaeo- 
zoological data. In Scythian and Greek settlements in addition to the bones or 
steppe animals (Saiga, Equus, Bos) the bones of forest animals (Cervus eaphus, 
Capreolus capreolus, Alces alces, Felix silvestris) were found. In rural settlements dated 
to AD bones of Putonus putonusL., which lives in oak forests were found^ 



settlement 



flavicollis Melchior 



com 



r 



part of the Dniepr Valley 



impoverished So io Slav and Tartar settlements (AD 1100-1200) Quercus 



254 



ABSTRACTS Vol. 12, No. 2 



dominate 



diminished after AD 1200 due to climatic 



human im 



Joseph E. LAFERRIERE, Arnold Arboretum 
Ave., Cambridge, MA 02138. 
POPULATION GROWTH AND THE ORK 
OF AGRICULTURE 






For any given state of cultural-environmental relations, food productivity is 
highest at medium population densities. Population growth would tend to 
increase beyond optimum levels but not to starvation densities. Populations would 
have a tendency to accept minor innovations or borrowings which might increase 
productivity in a short time frame but which may over a longer time period lead 
to population increases, thus creating an incentive for further cultural changes. 
Major changes which would result in large shifts in productivity would be 
discouraged because of short-term increases in labor costs. Changes which 
improve per capita productivity might result in a gradual improvement in lifestyles 
whereas changes requiring increased labor input would result in gradually decreas- 
ing per capita productivity. Genetic changes in resource species would have the 
same effect as technological innovations or shifts in time allocation. Temporary 
decreases in the productivity of a specific resource (e.g., crop failure due to 
drought) would increase the tendency toward acceptance of innovations, as would 
impending population decline due to depletion of nonrenewable resources. 
Depopulation due to war or epidemic might result in a shift toward less inten- 
sive agricultural systems. Resources amenable to enhancement of productivity 
would be favored over those for which opportunies for improvements are not 
evident. Diet breadth optimization theory predicts that cultural changes which 
result in increased productivity of resources previously neglected due to low 
efficiency might lead to broadening of the overall variability of the diet. Improve- 
ments in already preferred resources might reuslt in decreased diet breadth. The 
need for nutritional balance suggests an alternation of improvements in com- 



plimentary resources. 



Department of Anthropology, Smithsonian 



NMNH, MRC 112, Washingt 



SAMPLES FROM MARITIME 



LABRADOR 



samples have been collected from 



sam 



from 



8 by field crews of the Smith- 
William Fitzhugh. More than 
n dates run on these samples 



presents the Maritime 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 255 



ms 



found in the charcoal. It is shown that spruce was available for fuel during the 



Maritime 



Maritime 



environments 
)oen tundra ii 



Edelmira LINARES, Robert A. BYE, Jr., Jardin Botanico, Instituto de Biologia, 
Universidad Nacional Autonoma de Mexico, Apdo. Post. 70-614, D.F., 04510, 
Mexico, and Erick ESTRADA, Departamento de Fitotecnia, Universidad Auto- 
noma Chapingo, Chapingo, Mexico 
MEDICINAL VASCULAR PLANTS IN MEXICO: AN EVALUATION OF A 

BOTANICAL-CULTURAL RESOURCE 

Since prehispanic times, there have been inventories of medicinal plants in 
what is now known as Mexico. Thede la Cruz-Badiano Manuscript (263 plants listed 
of which 49% are identified), written by Aztecs, is one of the earliest written 
documents of this nature. Civil and church authorities of early colonial Mexico 
also collected data on remedial plants in the Natural History of New Spain (3,076 
plants listed of which 32% are identified) and the Florentine Codex (724 plants of 



Mexicano 



Medicinales (IMEPLAM) 



which was essentially based upon earlier results of the work of the Instituto Medico 
Nacional. IMEPLAM reported 2,196 higher plant species distributed in 900 genera 
and 161 families. Based upon our review of anthropological, botanical, 
ethnobotanical, and ethnomedical publications and unpublished theses since 1976, 
the Mexican medicinal flora registrv consists of 3,352 species (in 1,214 genera and 



families) 



only 24-60% of the data on medicinal employment 



Mexico 



me 



information 



vegetal remedies. Nonetheless, the lack of systematic comparative ethnobotanical 
data, of reliable taxonomic identification, and of field and laboratory validation 
of effectiveness indicates the urgent need for integrated surveys of medicinal 
plants, especially in light of continued loss of cultural information not passed onto 
the younger generation. Local studies as well as national surveys (such as those 



Mexicano 



will enable Mexico to evaluate the current status ot medicinal pan.. » 
ic manner and will aid the public and government officials inplannm 
maintenance and beneficial exploitation of this valuable cultural and 



botanical resource. 



Gary J. MARTIN, Department of Anthropology, University of California 
Berkeley, CA 94720 (94, Blvd. Flandrin, Paris 75116, France). 



256 



ABSTRACTS Vol. 12, No. 2 



WHO KNOWS?: AN EXPLORATION OF VARIATION IN MIXE 
BOTANICAL KNOWLEDGE 



I discuss an identification task aimed at measuring the variation in loca 
knowledge about plants in Totontepec, a Mixe community of Oaxaca, Mexico 
We prepared a set of 35 dried botanical specimens that exemplify the various us< 
categories, lifeforms, vegetational habitats, and other ways of characterizing plant: 
in Mixe ethnobotany. We then selected a subsample of 88 individials who repre 



mi 



munici 



specimens 



about plants. In this paper, I offer an analysis of how age and gender correlate 
with botanical expertise in Totontepec. 



Stephen MA VI and N.Z. NYAZEMA, National Herbarium and Department 
of Clinical Pharmacology, Harare, Zimbabwe 

RESEARCH IN MEDICINAL AND POISONOUS PLANTS: A CHALLENGE 
FOR THE HEALTH SYSTEM IN ZIMBABWE 



The aim of this paper is to highlight our current knowledge of the medicinal 
uses of plants in Zimbabwe and to give pointers as to the plants that need 
further investigations from chemical, medicinal, and pharmacological points of 
view. Five hundred traditional practitioners were visited in the main centers of 
Zimbabwe. Plants used by traditional practitioners together with information on 
their uses and how they were administered were recorded. Poisonous plants com- 
monly used by traditional practitioners were listed. Standard works published 
in Africa concerning traditional plant use were compared to our findings. Exten- 
sive studies have been carried out to investigate the medicinal properties of 
indigenous plants, their distribution, and cultivation. The 



aim 



conservation 



medicinal value. At the same time, effort has been made to encourage proper 
use of poisonous plants. These have included the following: Securidaca longepen- 
dunculata, Boophane disticha, Cassia abbreviata, Euphorbia ingens, Monadenium lugadiae, 
Warburgia salutaris, Gnidia kraussiana, Dioscorea sylivatica, Alepidea amatymbica, 
Solarium delagoense, Datura stramonium, Gloriosa superba, and Rauvolfia caffra. 
Research oq these plants may result in their utilization as either crude drugs or 
raw material in the manufacture of pharmaceuticals required to promote primary 



We 



much 



Guillermo Luis MENGONI 



LLAMA AS A MEAT 



logy, Smithsonian 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 257 






American camelids, both wild and domesticated 



lmoortant economic 



Although the domesticated camelids are better known as wool producers (especial- 
ly the alpaca, Lama pacos) and as beasts of burden (the llama, Lama glama), they 
were also an important source of primary products. Among them, meat and its 
related products should be considered as one of the main staples for many tradi- 
tional subsistence economies. 

This paper discusses the importance of the llama as a meat producer, based 
on the study of its economic anatomy. Available meat percentages and utility in- 
dices were calculated for the different body parts of the llama and compared with 
those of other ungulates. The information thus generated will be useful for 
having a better understanding of the present use given to this animal and the 
historv of its domestication. Certainlv, economic anatomy is another dimension 



animal domestication 



m 



and function. 



Pamela MCBRIDE, Julia E. HAMMETT, and Janet L. MCVICKAR 



Mexico, Albuquerque, NM 



CORN . . . FROM 



Little is known about the subsistence strategies of the prehistoric Sinaguan 



Mountains 



remains 



more 



removed from 



recovery 



of the large volume of well-preserved material obtained from this feature, we 
recognized a unique opportunity to make this comparison. The cache contents 
were subsampled and processed using a variety of commonly used macrobotanical 
extraction methods. A comparison of these results produced provocative infor- 
mation concerning the effectiveness and efficiency of each method. Though our 
results are specific to cinder matrix, similar results might be expected from other 
matrices. These methodological insights allow us to evaluate the precision ot eacn 

method. 



Mark D. MERLIN, General Science Department, Dean Hall Room 2, universiq 

of Hawaii at Manoa, Honolulu, HA 92822 

ETHNOBOTANY AND ENVIRONMENTAL IMPACT OF THE KAVA FLAW \i 
Piper methysticum (Forst. f.), ON POHNPEI ISLAND, FEDERATED SI Alt. 
OF MICRONESIA 



The 



me usycnoacuve K.UVU smuu^'r' » "-".?- — •<:;„„„* rolicnnim 

tropical Pacific, where it has been a major drug plant with fP^J "J"? 



medicinal 



most important plants in the Micronesian 



258 



ABSTRACTS Vol. 12, No. 2 



of Pohnpei, a high volcanic island in the Eastern Carolines. Traditionally, the root 
stock of this narcotic shrub has been a major offering of ceremonial tribute to 
high ranking individuals in the various districts of the island. Consumption of 
the kava beverage was also a traditional source of spiritual inspiration and a signifi- 
cant part of conflict resolution on Pohnpei. However, except for medicinal pur- 
poses, before European contact in the early nineteenth century, use of kava on 
the island was generally restricted to the chiefly class. During the twentieth 
century, the patterns of its cultivation and use on Pohnpei have changed 
dramatically. After colonial Japanese left the island, restrictions placed on local 
consumption were lifted, and farming and use of kava and its soporific beverage 
began to rise, with new segments of the society becoming involved. The rapidly 
increasing size of the human population, in combination with the nontraditional, 
widespread consumption of kava since WW II by both men and women at various 
levels of society, has put significant pressure on some traditional natural resource 
areas. Areas of native upland forest (nanwel) previously uncultivated are now 
subjected to ecosystem disturbance due to sometimes obvious, but more often 
clandestine patches of cultivated kava plants. This paper describes the evolution 
of contemporary patterns of kava consumption and cultivation. It also discusses 
the implications for long term soil erosion control and maintenance of biodiversi- 
ty among the native forest species of Pohnpei Island. 



MEZLUMIAN, Institute of Zoology, Academy 



375044, Armenia 



ARMENIA 



We have discussed the great number of horse remains in archaeological sites 
in Armenia from Eneolith to the Iron Age (fourth to first millennium B.C.) in 
previous publications. However, investigations of Bronze Age sites of northern 
Armenia have exceeded all expectations. Horse remains constitute about 90% of 
the domestic animals at these sites. Collections of subfossil equids from the 
foothills and mountain regions (former collections are from lowland sites) have 
been essentially replenished. Thus we have the opportunity to trace morpho- 
logical changes in the domesticated horse throughout Armenia. Based on cranio- 
logical parameters and the size and structural peculiarities of bone limbs, there 
appears to have been two separate horse populations in the Bronze Age in 
northern Armenia. At the same time there are appreciable differences between 
extreme variants. The most important in our opinion is the existence of the pro- 
nounced "breeds" from Dack-hor<?p<: fr> o-ra,w,,i o^.-n~ u~-™„ ,.,;«+, * withers 



cm 



m 



maintain 



mountain 



deposits of ore. Northern Armenia is just such a region. 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 259 



Christopher MILLER, Institute of Ecology, University of Georgia, Athens, GA 

30605 

BRAZIL NUT GATHERING IN EASTERN AMAZONIA: A SUSTAINABLE 

ALTERNATIVE TO DEFORESTATION? 



The Brazil nut tree (Bertholletia excelsa) is a forest canopy emergent in terra 
firme or upland sites in much of the Amazon Basin. Typically, 50-100 Brazil nut 
trees may be found together in stands or groves, called castanhais (sing, castanhal); 
groves are separated from each other by several kilometers. The trees bear fruits 
with 10-20 edible seeds, known as Brazil "nuts," that are gathered and sold by 
collectors. Most of the nuts are exported and sold in North American, European, 
and other markets. Brazil nut collecting in extractive reserves may generate 
revenues that surpass income from rubber tapping activities; furthermore, Brazil 
nut gathering may act as an important wet season activity to complement dry 
season rubber tapping in the forest. This paper will examine the yield and density 
of one hectare sample plots in several castanhais in eastern Para, Brazil and assess 
the values of the nut crops. Furthermore, priorities for future research goals on 
Brazil nut ecology and sustainability will be examined. 



NEWSOM, Department of Anthropology, Florida Museum 
. Museum Road, Gainesville, FL 32611, S. David WEBB, ai 



James 



FROM A FLORIDA WETSITE 



Recent excavations at the Page-Ladson (8JE591) site on the Aucilla River in 
Florida turned up Cucurbita pepo gourd seeds. Excavations of Page-Ladson have 
tested deposits that date from the last Wisconsin glacial phase (ca. 18,000 BP) 
to approximately 4000 BP. Lithic, bone, and wood artifacts from the Paleo-Indian 
and Archaic periods appear in late Pleistocene, early Holocene, and younger 

i i_ -pi •-. i •. i c u:~uu, ^r-r,^nir rlav <s«-rahim with 



come from 



Mamm 



human activity. One sample with gourd seeds has accelerator radiocarbon 



AMS lab, AA 



Thus 
more 



samples from 



from 



Steve PATTERSON, Department of Geography, 1255 Bunche Hall, UCLA, 

Los Angeles, CA 90024 

PHENOLOGY IN ECOLOGY AND ETHNOBIOLOGY 

Phenology is the study of periodic biological phenomena as they are £ fl « en «*| 
by climate and weather: the flowering of plants, the migration of birds, trie 

« .• c „^u nhpnnmpna are scat- 



emerg 



""^igciiLc ui insects, reports ui me uu^ivuuvm — ~- x m ^ 

tered throughout ethnographic literature, "The Omaha planted their squashes 



260 



ABSTRACTS Vol. 12, No. 2 



at the time of the blossoming of the wild plum" being an example. The obser- 
vation of phenological correlations can yield significant ecological information. 
Even in today's highly technological agriculture in the U.S., phenological data 
provides information as yet unmatched by any bio-climatic computer model. In 
Montana, for example, farmers make their first spring cutting of alfalfa within 
10 days after purple lilac begins to bloom, and thereby reduce their losses to the 
alfalfa weevil, Hypera postica, which emerges at that time. 

Phenological observations also become incorporated in noneconomic cultural 
patterns, for example, in the timing of ritual occasions. According to Radcliffe- 
Brown, the Andamanese had a calendar of scents, based on the succession of 



came 



coming 



time 



harvest 



link 



systemat 



r 



phenology, and highlights questions and problems 
istances of such questions which emerged during 



ritual and ornamental plants among the Yucatacan Maya are adduced as examples 

QUAN, Xia, Department of Biology, Lanzhou University, Lanzhou, Gansu 

Province, P.R. China 

A STUDY OF GANNAN TIBETAN'S TRADITONAL MEDICINE: HERBS 

AND HEALINGS 

Gannan Tibetans live in the northwestern part of China. For centuries, tradi- 
tional Tibetan medicine has played an important role in the health care of the 
Tibetan communities of the region. Today, Tibetan medicine is still practiced as 
a parallel system, or in some isolated regions, as the dominant medicinal system 
compared to the modern medicinal system. In the medicine of the Gannan 
Tibetans, a number of highland herbs are employed. The methods used in pro- 
cessing and utilizing these plants are quite different from modern pharmacological 
methods of treatment. Some of the herbal medicines are very valuable in the 
development of cancer-killers or even AIDS-killers. In this paper, the author 
reports the results of a study of the medicinal system of the Gannan Tibetans. 
One hundred and fifty species of traditional herbal medicines are identified; their 
scientific names, local names, usages, and parts used are recorded. Botanical and 
pharmacological aspects of some plants are briefly discussed. 

Gannan Tibetan Autonomous Profecture is located in the convergence of the 
Qingzang plateau and Loess-highland of China. With population of 5.5 million, 
the autonomous profecture is a major Tibetan region of Gansu Province. Tradi- 
tional Tibetan medicine has a history over 1000 years old. Many centuries before 

the Christian era. thp anrpstnrs: of Tihotanc Vr»o™, fV,^>t A\ c t,\\aA Viicrhland barley 



grains could cure injury 



medicine, traditional Tibetan herbal medicines 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 261 



important role in the indigenous me 
lannan Tibetans. In a Tibetan commu 



mainlv in Buddhist monasteries 



maintain the precious cultural heritage, many 



armaco 



medicine. These individuals have accumulated 
d herbal medicines used in the treatment of ill 



most 



mvsticism" of Tibetan Buddhism 



com 



Gannan 



medicine svstem 



GA 30602 



J. REITZ, Museum 



THE SIZE OF DOMESTIC CATTLE IN COLONIAL SITES 

A significant aspect of early colonial efforts was the adaptive response in 

m ^ _ _ ^ ^ — ^ % * i Jt _ __ ■ . L ^ ^ LL I j-*. 



remains. Measurements 



from Caribbean Hispanic 



American 



However 



which mieht be involved include the harsher climates 



com 



origins of livestock found at Spanish and English North American colonies. 

Irwin ROVNER, Binary Analytical Consultants, North Carolina State University 
Raleigh, NC 



PHYTOLITHS 



HISTORIC PERIOD 



em 



onmental 



studies 



more 



much of ohvtolith deposition allows for the study of micro 



mental and micro 



_ Recent studies at a number of 

Montfcdlo (VA), early Hampton (VA), Harpers Ferry 
(WV), among others, demonstrate a high degree of precision in identifying 
ethnobotanical activity. These include alteration of local floral assemblages, 



patter 



history 



>anning its most recent two centuries. This reconstructic 
documented history of the contiguous house and its occu 
i recovered from tartar deposits adhering to domestic her 



262 



ABSTRACTS Vol. 12, No. 2 



teeth recovered from excavations at early Hampton, VA. Distinct differences in 
animal dietary patterns contributed to greater awareness of land use and live- 
stock management practices at both the household and community levels over 
a period of some 150 years. Microfossil data captured in the teeth appear also 
to provide evidence of local water quality, e.g., increased degradation and pollu- 
tion, over time. 



Irwin ROVNER, Binary Analytical Con 
Raleigh, NC 

PROBLEMS OF MORPHOMETRIC P 
IDENTIFICATION AND ANALYSIS 



Size and shape measurements, morphometri 
seed identification manuals and atlases, are t) 
measurements using small— often unstated— sam 



few 



computer-assisted ima 



multiple measurements of size as 
sample populations for each taxon, respectively. Included were measures of size 
(length, width, breadth, area, and perimeter) and shape (aspect ratio, form factor, 
roundness, extent, solidity, and convexity). Comparison of morphometric data 
of the populations tested with data presented in standard references displayed 
substantial discrepancies. For instance, normal range of size variation often 
exceeded, even far exceeded, published data. Discrepancies occurred 
unpredictably— sometimes at the minimum end, sometimes at the maximum end, 
sometimes at both ends of the range. In a few instances the ranges were so dif- 
ferent that no overlap of size occurred. In many cases the mean size value for 
the large sample population fell outside the range presented in standard 
references. This means that more than half of a oomilation of seeds selected at 



measure 



Similarly, comparison of seed populations at the varietal level showed very distinct 
ranges of variation; some narrow, some broad. Thus, if seeds of one variety are 
compared with a reference population of another variety within the same species, 
a considerable disparity of size is possible, if not probable. 

Typically, archaeobotanical and paleobotanical seed populations are randomly 
selected, i.e., randomly deposited and recovered. This suggests that the same 
discrepancies noted above between the large tested populations and standard 
small population reference data may also operate in the analysis and interpreta- 
tion of fossil seed populations. This raises some doubt, for example, about the 
interpretation of seeds as cultivated due to "increased" size if comparisons are 



made 



morphometric 



assumes a normal distribution, may 



parametric 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 263 



Jan SALICK, Department of Botany, Ohio University, Athens, OH 45710 
NON-TIMBER FOREST PRODUCTS, ETHNOBOTANICS, AND BUFFER 
ZONE MANAGEMENT IN THE INTERNATIONAL PEACE PARK, 
NICARAGUA AND COSTA RICA 



nontimber 



comoonent of extractive reserves 



nagement. A diverse array of species can be extracted from tropical rainforests 



being attempted 



Management 



remarkably preserved throughout the war, only to be battered even more furiously 
now by laissez-faire economic extraction and land grabbing. Buffer zone manage- 
ment holds some faint hope for conservation and protection of the inner preserve. 
Experiments and field data compare community characteristics of nontimber forest 
products under natural forest management, campesino management, and primary 
and secondary forest conditions. Campesino management appears to be the next 
best thing to primary forest as far as species diversity, and best yet as far as 
ethnobotanics. 



Enrique SALMON, Arizona State University, Tempe, (1632 East Northshore t 
Tempe, AZ 85203) 

EXAMINING THE TARAHUMARA PERSPECTIVE OF SACRAMENTAL 
CORN BEER: TOWARDS AN ETHNOGRAPHIC RE-CLASSIFICATION 
Zea mays 



e Tarahumara of Chihuahua, Mexico, corn is their staff of life. It is be 
Tarahumaras that thev are the children of corn. As a result, they em 



as a sacramental offering, corn beer, known within the culture as suwi-ki. The 
beer is derived from the plant Zea mays, along with other native Mexican plants. 
Past ethnographers have misclassified the corn beer made from Zea mays as only 
a beverage and special food. However, John Kennedy's etic approach considered 
the beer within its religous context, but fell short of describing its true character. 
In his book on Tarahumara foods, Rela'muli Nu'tugala Go'ame (Food of the Tara- 
humaras), Albino M. Trias, a Tarahumara, emically does not include suuri-ki, 
except to mention how it is made, thus implvine that the Tarahumara classify 



mething 



Tarahumara 



emic view 



ment." The Tarahumara ritual observance of the bevei 
demonstrating its unique ceremonial and communal 



Judith SCHMIDT, New York Botanical Garden, Bronx, NY 
PLANT DYES FOR HARRIS TWEEDS 



264 



ABSTRACTS Vol. 12, No. 2 



What are the plant dyes traditionally used for Harris tweeds in the windblown 
North Atlantic islands of the Outer Hebrides off the west coast of Scotland? Are 
the same dye plants used on the mainland? 



Elizabeth M. SCOTT, Mackinac State Historic Parks 
AT HOME THEY 'COMMONLY ATE NO PORK': OBSERVANCE OF 
JEWISH DIETARY RESTRICTIONS AT THE SOLOMON-LEVY TRADING 
HOUSE AT FORT MICHILIMACKINAC, 1761-1781 



This paper examines the evidence for adherence to Jewish dietary restrictions 
in one eighteenth century household in North America. Land sale records from 
the fur trading post of Fort Michilimackinac, in what is now northern Michigan, 
identify this house as one purchased in 1765 by two Jewish traders, Ezekiel 
Solomon and Gershon Levy. According to both the material culture and dietary 
evidence, the occupants of this house between 1761 and 1781 became increasingly 
wealthy. Documentary evidence for Solomon reveals his increasing wealth in the 
1770s and his increasing involvement with the Jewish congregation in Montreal. 
The change in diet evident in the food remains corresponds to both factors, and 
it is the dietary evidence rather than the material culture which suggests the 
Jewishness of the occupants. 

The paper presents evidence from faunal and floral remains for this household, 
as well as from documentary records. By comparison with contemporaneous 
neighboring households, the diet in the Solomon-Levy house is shown to be 
distinct from that of both British and French-Canadian colonists. Following F. 
Gerard Ijzereef's (1989) criteria for identifying Jewish and nonjewish households 
in eighteenth century Amsterdam, the evidence from the Solomon-Levy house 
at Michilimackinac may be seen to indicate greater adherence to Jewish dietary 
restrictions through time. 



G.K. SHARMA, Biology Department, University of Tennessee, Martin, TN 
38238-5014 

Cannabis STUDIES IN THE HIMALAYAS 



The Himalayas are known for their wild or nearly wild populations of Can- 
nabis. Botanical studies of Cannabis populations in their natural habitats in the 
Himalayas are of great significance for understanding the ecological, biochemical, 
and ethnomedical relationships existii 
certed, thorough investigation of this 
home in the Himalayas. The finding 
Himalayas will be discussed. 



mysterious plant be made 



,bable 



Medicine, Beijing, 100700 
! ADVANCES OF CHI1 



Materia Medica, China Academy 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 265 



This paper deals with the advances of Chinese ethnoherbology, stressing the 
two aspects of traditional and multitribal characteristics. On one hand, the "tradi- 
tion" of a science means that it has lasted for many centuries, beginning in 
ancient ages, is rooted in the thoughts of the population, and has been used for 
a long time. On the other hand, Chinese ethnoherbology is an integration of all 
ethnic medical systems. In other words, it contains various constituents of Tibetan, 
Mongolian, Uygur, Korean, and other minority systems in addition to the Chinese 
system which is the main part of China ethnoherbology and already has a great 
reputation all over the world. There are five points reviewed in this paper: 
(1) a new term "ethnoherbology" first appears in this paper; it can be defined 
as the scientific study of the relationship of ethnology and herbology; (2) the main 
systems of China ethnoherbology include Chinese, Tibetan, Mongolian, Uygur, 
and Korean, which belong to their own language families, necessitating the study 
of the relationship between languages and ethnoherbological systems; (3) the 
results of exploration and survey of Chinese ethnoherbology indicate that there 
are at least 2,000 ethnomedicinal items, commonly used by the ethnic healers, 
and that these are new medicinal resources to be developed for health care of 
modern society; (4) comparative ethnoherbology is, and will be, an attractive sub- 
ject which will benefit both the discovery of new drugs and the study of the rela- 
tionship among tribes; (5) China ethnoherbology has been, and will be, successful 
in developing more powerful remedies. 



Ashok Kumar SRIVASTAVA, Department of Botany, MG. Degree College, 
Gorakhpur 273001, India 

THREATENED MEDICINAL PLANTS OF TARAI FOREST OF GORAKHPUR 



Increasing demands for indigenous herbal drugs and their exploitation by 
aboriginal tribes, traders, researchers, and recently by pharmaceutical companies 
threaten our wealth of valuable medicinal plants. Urbanization and unplanned 
growth of residential areas also affect plant distributions. This paper highlights 
some important plants of the Gorakhpur forest which need urgent conservation 
before they become extinct. The important plants of this area are Aristolochia 
indica L., Asparagus racemosus Wild, Dioscorea bulbifera L., Flacourtia jangomas (Lour.) 
Keeusch., Gloriosa superba L., Helminthostachys zeylanica L., Ophioglossum reticulatum 
L Rauvolfia serpentina Benth. & Kurz. and/?, tetraphylla L. These are described 
along with parts used, habitats, localities, and causes of threat. 



Autonoma de Mexico 



Apartado Postal 70-275, Mexico, DF-04510, Mexico 

GREEN ECONOMICS AND INDIGENOUS WISDOM: 

HOW MANY PRODUCTS ARE ENCLOSED IN A TROPICAL FOREST? 

he whole spectrum of forest products recognized by eight indigenous groups 
ot tropical Mexico is presented and analyzed. More than 2,500 "ethnoproducts" 
corresponding to over 1,500 plant species were identified from the tropical rain 



266 



ABSTRACTS Vol. 12, No. 2 



forests of Mexico. For each species information is given about scientific and 
common name, botanical family, location, type of use, part used, lif ef orm, habitat 
and indigenous groups from which the data were obtained. The number and 
quality of these products are also reviewed on four different scales (sites, localities, 
microregions, and the entire nation). The results of this research are then com- 
pared with those obtained by other authors in different tropical forests of Africa 
and Latin America. The paper ends by discussing the importance and limitations 
of this tropical indigenous wisdom for the development of a new green economy. 

Mollie S. TOLL, Office of Archaeological Studies, Museum of New Mexico, 



NM 



WAREROOM 



POST: CHRONICLING SOME ECONOMIC PATTERNS OF NAVAJOS 
HOPIS, AND ANGLOS IN THE LAST 100 YEARS 



Hubbell 



Park Service, recently underwent some 



wareroom 



nical imports. Archival photos and inventories provide specif i< 
materials recovered, fleshing out the list of materials passinj 



wareroom 



bance, and archaeological collection. The growing body of archaeobotanical data 
sets from other trading posts in the Four Corners area, and from historic sites 
on the Hopi, Navajo, and Zuni reservations, allows views into the interplay o 
indigenous and introduced domesticated crops. 

Michael TOPLYN, Harvard University (15 Highland Ave., Somerville, MA 02143) 
LIVESTOCK, LIMIT ANEI, AND STRATEGIES OF MILITARY SUBSISTENCE 
ON ROME'S ARABIAN FRONTIER (A.D. 284-551) 

Although the past 30 years have seen an upsurge of archaeological interes 
in the lifestyle, deployment, and maintenance of Roman soldiery, there has been 
little emphasis on understanding the subsistence economy of the frontier legions, 
a research problem which can be addressed directly through the study of too 
refuse. To date, most zooarchaeological insights bearing on the logistics of foo 
provisioning for the Roman army have been based on studies of skeletal materia 
from excavated sites in Europe. Comparable evidence for the dietary maintenance 
of Rome's military establishment in Arabia is practically nonexistent. 

This paper represents the first attempt to fill that vacuum by merging the 
results of archaeofaunal analysis with ancient literary sources. Specifically, the 
large sample of animal bones (ca. 85,000 fragments) recovered from military sites 
of the Limes Arabicus Project in central Jordan (el-Lejjun, Rujm Beni Yasser, 
Qasr Bshir, De'janiya) is used to investigate the economic status of the limitanei, 
or frontier troops of Roman Arabia. My research involves intersite comparisons 
of mortality profiles, carcass part distributions, and osteometric results. At issue 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 267 



is the long-lived controversy centered on the functions of the limitanei, who once 
were thought to have evolved, by the fourth century, into a peasant militia of 
part-time soldiers which farmed government lands and performed the duties of 
frontier guards. Current historical opinion, relying solely on literary evidence, 
favors the view that frontier garrisons were largely reliant on the imperial 
bureaucracy for imported food. The methodology I have adopted to address this 
problem involved the identification of livestock culling practices through the 
application of skeletal ageing techniques (epiphyseal fusion, tooth eruption/ wear). 
Mortality profiles for sheep and goats are contrasted with ethnographically derived 
models of caprine production which differentiate between animal exploitation for 
local subsistence and herd management for production goals geared to exchange. 
The archaeofaunal evidence of the Limes Arabicus Project clearly indicates that 
the pastoral economies of frontier garrisons were characterized by subsistence 
autonomy in the late third through mid-sixth centuries. Bones of domestic 
caprines consistently reflect patterns of exploitation typical of the agro-pastoral 
economies of peasant farming communities. Temporal distributions of caprine 
carcass parts support mortality data in suggesting that military sites neither 
obtained meat from outlying settlements, nor routinely supplied it to military or 
civilian sites. However, the bones of pack animals (donkeys, mules, camels) 
imply the regular importation of supplies to frontier outposts, supplies which 
included foodstuffs such as parrotfishes (Scaridae) which could not be acquired 
locally. Moreover, the use of camel hybrids in overland transport is suggested 
by results of osteometric analysis. Although the zooarchaeological evidence of 
the Limes Arabicus Project does not conclusively prove that all limitanei stationed 
on the eastern frontier were peasant farmers, it does demonstrate a far greater 
degree of subsistence autonomy for frontier garrisons than generally has been 
supposed. This result has profound implications for an expanded understanding 

of Roman frontier policy and military recruitment practices in the time of the Late 
Empire. 



Julian TREVINO-VILLARREAL, L. CORRAL-PEREZ, A.M. DEi 
GUILLEN, R.I. JIMINEZ-SILVA, J.A. MALDONADO-HERNAr 
J.A. BALDERRAMA-ALARCON, Instituto de Ecoloeia v Alimentos 



ma 



Tamaulipas, 87040 



Tamaulipas, Blvd. Adolf o Lopez Mateos 928, CD Victoria, 



5TUDY OF MAMMALS FROM 
TAMAULIPAS, MEXICO 



G i^ 1! R StUdy P resents information about the utilization of mammals at "El 

leio Biosphere Reserve, Tamaulipas, Mexico. The objective of this research 

as the elaboration of checklists that would increase our knowledge of the kinds 

mammals utilized by local people and how these mammals were used by them. 

ree hundred and twenty one interviews were carried out in 13 different local 

1 ill m \ ^^ "t ^ *<^ - 1 « _ ^ 



the areas was made 



album of mammals found in 



268 



ABSTRACTS Vol. 12, No. 2 



computer program 



Our results showed that only 49.5% of the local people utilized mammals 



in some way. These 



mammals. The species most utilized were the nine-banded Armadillo 
"(Dasypus novemcinctus) and the eastern cottontail (Sylvilagus floridanus), at 10.1% 
each The least utilized were the southern flying squirrel (Glaucomys volans) and 



main methods 



me 



discusses the different ways mammals are utilized by the local communities. 

Chusie TRISONTHI, Department of Biology, Faculty of Science, ChiangMai 

University, ChiangMai, 50002, Thailand 

EDIBLE WILD FRUITS IN NORTHERN THAILAND 



Thailand stretches from the wet tropics north into the dry subtropical region, 
where plant species diversity is high. The wild plants in northern Thailand are 
of tremendous importance to the rural economy. Edible fruit is one of the impor- 
tant forest products. These fruits are collected from the forest and sold in the 
markets. Some of these fruit plants are becoming endangered because ecological 
problems, such as drought, affect their growth. 

A survey of wild plants producing edible fruits was carried out with the 
purpose of increasing public awareness of their conservation. Over 50 species 
of wild plants producing edible fruits from eight provinces of northern Thailand 
were collected and identified. These fruit plants belong to 40 genera and 25 
families. The dominant species, such as Canarium subulatum, Castanopsis dwersifoha, 
Irvingia Malayana, Protium serratum, Schleichera oleosa, and Spondias pinnata are 
distributed in all provinces. Details on each species, such as local names, scien- 
tific name, location, and method of preparation for food are fully described in 
the presentation. 

Nancy J. TURNER and Alison DAVIS, Environmental Studies Program, Box 3045, 
University of Victoria, Victoria, B.C., V8X 3P4, Canada 
"WHEN EVERYTHING WAS SCARCE:" THE ROLE OF PLANTS AS 
FAMINE FOODS IN NORTHWESTERN NORTH AMERICA 

Interviews with indigenous elders and literature reports indicate that p 
have played an important role as survival foods for indigenous peop es 
northwestern North America. Over 50 species of food plants are noted specitica y 
to have been used to alleviate hunger and aid in survival in lean times, in P 
historic and early historic -times, food shortages occurred periodically, usua y 
late winter and early spring when bad weather and other circumstances c 
cided with low quantities of stored foods and unexpected scarcity of fish and g 
These famine foods were also resorted to by non- Aboriginal explorers, tra ' 



Most 



with 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 269 



some marine algae, the lichen Bryoria fremontii, tree species having edible inner 
bark (e.g., Pinus contorta, Tsuga heretophylla) , species of ''root" vegetables (e.g., 
Camassia spp., Trifolium wormskioldii, Pteridium aquilinum, Lysichitum americanum, 
Dryopteris expansa, Claytonia lanceolata, Cirsium undulatum), green stem and leaf 
vegetables (e.g., Opuntia spp., Heracleum lanatum), and long-lasting fruits such 
as Arctostaphylos uva-ursi, Viburnum edule, and Rosa spp. As well as famine foods, 
some plants (Polypodium glycyrrhiza rhizomes, Gaultheria shallon leaves, Blechnum 

spicant fronds) were used as temporary hunger suppressants by hunters or 
travellers lacking more substantial foods. The dietary contributions of these plants 
are probably variable, but all should be regarded as highly significant because 
of their survival value. 



Gail E. WAGNER, Department of Anthropology, University of South Carolina, 

Columbia, SC 29208 

NATIVE AMERICAN GARDENS IN EASTERN NORTH AMERICA 



There is little direct archaeological evidence on Native American gardens and 
landscaping in eastern North America, although we have a growing data base 
on what plants were used or grown. Spanish, French, and English accounts and 
drawings from the sixteenth through the eighteenth centuries provide scattered 
descriptions of fields, gardens, and other maintained landscapes. In this sum- 
mary I examine not only what crops were grown or encouraged, but also where 
they were placed and who managed them. 



David E. WILLIAMS, National Germplasm Resources Laboratory, USDA-ARS, 
Building 003, 4th Floor, BARC-West, Beltsville, MD 20705 
IMPACT OF NATIVE FARMERS ON PEANUT DIVERSITY: PAST, PRESENT, 
AND FUTURE 



lm 



of peanut diversity has not been fully appreciated. Very little direct evidence 
remains, other than the peanuts themselves, to indicate the exact means by which 
such great diversity evolved in prehistoric times, but there is a clear correlation 
between peanut geno-centers and the aboriginal culture areas. Presently, native 
farmers throughout the warmer parts of the Americas continue to maintain a great 
deal of peanut genetic diversity as part of their ancestral farming systems. A 



some 



instrumental in generating and maintaining 



becomes 



farmers neec 
maintenance 



270 



ABSTRACTS Vol. 12, No. 2 



Elizabeth S. WING, Florida Museum of Natural History, University of Florida, 
Museum Road, Gainesville, FL 32611, and Stephen R. WING, University of 
California-Davis 
COLONIZATION OF ISLANDS 



In their colonization of islands, pre-industrial people faced limitations 
:ribed by island biogeographical models. Two ways in which these limitations 
e mitigated were by the introduction of domestic and captive mainland animals 



exploitation of marine resources. Many examples of p 
animals to islands bv DeoDle have been documented 



times 



vated from sites on islands in the Caribbean demonstrate the exploit; 
number of different marine habitats. We used measures of species dive 
equitability to examine the degree to which human technology can com 
for the limited diversity of terrestrial fauna on islands that are small 
ant from the mainland. 



L. Anthony ZALUCHA, Paleoethnobotanical Consulting, 109 Sunset Lane, Mount 
Horeb, WI 53572 

WOOD CHARCOAL FROM THE VALLEY VIEW SITE 



The Valley View site (47LC34), located near La Crosse, Wisconsin, is an 
Orr Phase Oneota village occuped about A.D. 1500. The intensively excavated 
northern portion of the site revealed a complex of pit features. Intensive matrix 
sampling from the features resulted in a large data base of charcoal from discrete 
depositional episodes. Five morphological feature types were defined. Different 
patterns of species presence were noted across the feature types. Statistical analysis 
suggests that these differences are due to cultural causes, not to chance factors 



number 



time 



Xeric forests and conifer swamps were the preferred wc_ 
from the latter zone must have been highly prized since v 
tion shows that the nearest swamps were several miles 
burned structure shows that oak, birch, and tamarack wen 



Woods 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 271 



BOOK REVIEW 



The Abandoned Narcotic: Kava and Cultural Instability in Melanesia. Ron Brun- 
ton. Cambridge Studies in Social Athropology, 9. Cambridge, U.K.: Cam- 
bridge University Press, 1989. Pp. viii, 216. $39.50. ISBN 0-521-37375-1. 



Prior to the entry of Europeans into the Pacific, relatively few plants were 
used by islanders as stimulants or drugs. One of these, Piper methysticum, was 
used in the preparation of kava, a drink that everywhere was prepared by chew- 
ing, grating, or pounding the roots of P. methysticum and mixing the results with 
water; no fermentation was involved. The plants' pharmacological properties 
clearly indicate its potential as a soporific, anticonvulsant, muscle relaxant, and 
local anesthetic— precisely the effects sought by kava drinkers, especially as they 
are seen to facilitate ritualized interaction with the supernatural world. 

Brunton is concerned primarily with accounting for the geographical distribu- 
tion of traditional kava drinking, which is indeed curious. At the time of Euro- 
pean contact, kava was a ritually important drink throughout Polynesia, but 
only on two islands in Micronesia and in a few, widely separated parts of 
Melanesia. W. H. R. Rivers, one of the major proponents of diffusionist ap- 
proaches to Oceanic culture history, argued in 1914 that such a distribution 
reflected the historical movements of two waves of immigrants to Melanesia from 
Southeast Asia, "kava people" and, later, "betel people," i.e., those who 
chewed betel nut (Areca catechu) with the leaves of Piper betle and slaked lime. 
Rivers proposed that kava, which was rarely plentiful and involved prolonged 
preparation, was abandoned by most Melanesians in favor of the simpler pro- 
cess of chewing betel, which allegedly induces comparable effects. Marshalling 
the available archaeological, botanical, ethnological, and linguistic evidence, Brun- 
ton explores Rivers's general notion that kava drinking was formerly more 
widespread in Melanesia and subsequently abandoned in most locations, as well 
as a competing hypothesis of multiple independent inventions. 

The diverse and complementary evidence Brunton assembles is persuasive, 
and it would indeed seem that "there can be little doubt that kava drinking had 
a single point of origin" (p. 75). It also appears that precontact direct links 
between the kava-drinking areas of Melanesia are "totally implausible" (p. 78), 
leading to Brunton's conclusion that the distribution of kava drinking in Melanasia 
at the time of European contact is to be explained by postulating that "there were 
people in intermediary areas . . . who once drank kava as well, but they aban- 
doned it some time before European contact" (p. 80). The intriguing remaining 
issue, then, has to do with why this abandonment took place. 

Brunton rejects Rivers's argument that kava drinking was abandoned in favor 
of betel chewing and I think he is correct to do so, if for no other reason than 
that the two drugs are not necessarily competitors, and indeed they do not have 
mutually exclusive distributions. For any given location, Piper methysticum's 
requirements of high rainfall, well-drained soil, and shelter from sun and wind, 
combined with its apparent dependence on vegetative propagation made it 



272 



BOOK REVIEW Vol. 12, No. 2 



likely that a climatic catastrophe could make abandonment an irreversible process 
(pp. 90-92), and perhaps no reasons other than ecological ones need be sought. 

T% ■ n ■ • • i . i it. i y i • s s . « _• ii 



him "the most im 



most 



/ / 



for kava's abandonment 
Melanesia" (dp. 92, 93). 



almost 



exemplify processes 



and continue to be widespread in Melanesia. He supplements his Tannese material 
with citations of the ethnographic literature to make to general points: (1) that 
"the major significance of kava in virtually all the societies in which it was drunk 
was ritual or religious" (p. 170); and (2) that Melanesia, and particularly New 
Guinea, is rife with examples of major ritual complexes or practices being modified 
or abandoned in precontact times or following European contact and, according 
to Brunton, for reasons similar to those behind kava's varying fate in Tanna, that 
the drug (or cult, or ritual) was perceived to be either "too strong" to be safe, 
or the opposite, that is was no longer efficacious in the light of social changes. 
Why, then, was kava drinking retained in some Darts of Melanesia, either 



the 



abandoned? Brunton proposes 



communities 



Melanesian specialists 

v variable (and much r 



in Polynesia, where kava 



social orders there), but a systematic mapping of this variation against the distribu- 
tion of kava drinking is the obvious next step needed. 

Brunton stops short of such a project, but his innovative work will doubtless 
stimulate others to attempt it. And that will surely be a breath of fresh air in 
today's "postmodern" anthroooloev. whirh 



seems 



especially welcom 



our attention to the importance of attending to nonfood plants as we continue 
to explore the boundless relevance of ethnobinloav to n„r hrn^Hpr understanding 



human 



Terence E. Hays 
Rhode Island College 
Providence, RI 02908 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 273 



NEWS AND COMMENTS 



PROJECTS AND PROGRAMS 



The Central American Institute of Prehistoric and Traditional Cultures at Belize 

This institute was established under a charter from the government of Belize 
as a scientific, educational, and nonprofit organization for the purposes of pro- 
moting the preservation of ancient and traditional cultural ethos and materials, 
and to act as a center for the dissemination of knowledge and interest in the study 
of such cultures. These activities include scientific research and archaeological 
investigations, preservation for public benefit of archaeological monuments and 
historical landmarks, as well as academic and educational programs on all aspects 
relating to prehistory, ethnohistory, and the ethnographic present. 

A total of 27 acres of jungle terrain, kept as a natural preserve, have been 
dedicated for an ethnobotanical field station, which is to serve as a research facility 
where plants of ethnomedical interest and ritual importance can be studied in 
their native habitat. The Institute also encourages work in wildlife and tropical 
resource management, to be pursued from indigenous and traditional points of 
view. The Institute's address in Belize is 20 Macaw Avenue, Belmopan, Belize, 

Central America. 

For more information contact Dr. Michael Ripinsky-Naxon, Director, Central 
American Institute, 68-769 First Street, Suite 286, Cathedral City, CA 92234-1244, 
USA; fax number 1/619/324-8862. [Extracted from a press release and other 
materials submitted by Agata Dukat, administrator of the Institute]. 



JOURNALS AND OTHER MEDIA 



Etnoecologica, a new journal edited by Victor Toledo of the Center of Ecology, Na- 



1992 



Autonomous University of Mexico 
management and conservation of n 



dated April, 



contains 



commentaries; breves, which are short communications; debate, includin 



and libros closes the journal with book reviews. 



from indie 



are 



udes two numbers, is available from: Victor M. 
Ma. Guido, Morelia, Michoacan 58090, Mexico 



5259. 



COMMENTS, OPINIONS, AND RESPONSES 



At present, the Journal of Ethnobiology accepts scientific articles, short com 
munications, book reviews and, for this column, newsworthy items, and com 
ments on papers published in the Journal In a slight departure from this policy 



274 



NEWS 



Vol. 12, No. 2 



Willard Van Asdall, the former editor of the Journal, accepted an article by Dar- 
rell Posey entitled "Intellectual property rights: what is the position of 



ummer 



submitted to reviewers, addressed some 



economic lm 



Following 



Comments column. Readers are invited to submit 



pieces, of 500 words or less, which address the applied aspects of ethnobiological 
research or other issues of concern to ethnobiologists. [By the editor of News and 
Comments] . 



REQUESTS FOR INFORMATION 

An issue on the minds of some participants at the Fifteenth Annual 
Ethnobiology Conference in Washington, D.C., was the lack of information on 
university-level degree programs and courses on ethnobiology. Ethnobiologists 
often receive requests from students about where to seek advanced training in 
our field, and are unable to give a suitable answer. Eugene Hunn has suggested 
that a directory be published, perhaps in the Journal, of professors and academic 
institutions that offer training in ethnobiological theory and methods. If you know 
of programs, whether in the United States or in a foreign country, please con- 
tact the editor of the News and Comments section. After reviewing the initial 
responses, I may conduct a more formal survey and publish the results in the 
Journal. [By the editor of News and Comments]. 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 275 






BOOK REVIEW 



Early Animal Domestication and its Cultural Context. Pam J. Crabtree, Douglas 
Campana, and Kathleen Ryan, editors. Philadelphia: MASCA Research 
Papers in Science and Archaeology, Special Supplement to Volume 6; The 
Museum of Applied Science Center for Archaeology, The University Museum 
of Archaeology and Anthropology, University of Pennsylvania, 1989. Pp. 
vii, 136. 54 figures; 19 tables; no index. $32.00 U.S. (hardcover). ISSN 1048- 
5325. 



This slim, well-edited volume is a collection of separate papers put together 
as a tribute to Dexter Perkins. Perkins (1927-1983) and his wife Patricia Daly were 
in the forefront of research on the origins and development of animal domestica- 
tion in the Near East during the 1960s and 1970s. 

A short biography of Perkins by R. Solecki and review of Perkins's scholarly 
contributions by Crabtree and Campana introduce the book. The remaining seven 
papers are accounts of original zooarchaeological research, several of which deal 
with samples originally described by Perkins, and others relevant to the Near 
East or the topic of animal domestication. Three papers focus on analytical 
problems: the distribution of wild sheep in the Indus valley (R. Meadow); a 
reappraisal of a histological screening technique proposed by Perkins and his 
colleagues for the determination of domesticated status in animal bone (A. 
Gilbert); and an overview of the archaeology of horse riding (D. Anthony and 
D. Brown). The Gilbert paper, the longest in the book, reveals some of the 
development of zooarchaeology as a science. Patterns in bone thin sections had 
been interpreted as signs of domesticated status, but now appear to be diagenetic 
artifacts, a familiar disappointment to many who work on degraded archaeological 
samples. Patterns in the architecture of spongy bone may indeed indicate some 
difference in wild and domestic activity patterns, as Drew, Perkins, and Daly had 
suggested, but establishing a modern baseline for such changes has been pro- 
hibitively time consuming and expensive. This also is a familiar story to those 
working on the biological effects of human control of living animals. 

The remaining papers focus of assemblages of animal bones in a regional 
context. The authors cover early animal domestication in India (P. Rissman); the 
middle and upper Paleolithic of the Zagros (B. Hesse); hunting in early village 
economies in Anatolia (G. Stein); and the colonization of northern Europe (P. 
Bogucki). Taken together, these papers show how vast an amount of site-specific 
data must be collected before interesting cultural patterns emerge. The collective 
weakness of these papers is the lack of coordination between analyses of animal 
remains and analyses of forage and crops, partly, but not wholly, because the 
latter data do not exist. 

One of Perkins's accomplishments was his insistence on the place of zoo- 
archaeology within the field of archaeology, rather than zoology as had previously 
been the tradition in the United States. Here, the authors deal with chronologies, 



276 



BOOK REVIEW Vol. 12, No. 2 



assemblages as thev interpret patterns in animal 



assemblages. At the same time 



many physical sciences in addition to mammalian morphology and bone biology. 
The fruits of this approach are a reconstruction of animal use in the Near East 
which is deliberately focused upon human use of, and alteration of, the natural 
environment. This readable book is not comprehensive enough in coverage to 
be used as a basic reference for this region, but it is a showcase for research 
methodology and regional synthesis on an important problem and area. 



Katherine M. Moore 
Department of Behavioral Science 

Bentley College 
Waltham, MA 02154 



BOOK REVIEW 



The Tlingit Indians. George Emmons, edited with additions by Frederica de 
Laguna. Biography by Jean Low. Seattle and London: University of 
Washington Press, 1992. Pp. 530. 65 line drawings, 127 photographs. $60.00 
(hardcover). ISBN 0-295-97008-1. 



is 



The Tlingit Indians is sumptuously produced, encyclopedic, and well il 
strated. It represents a monumental effort of some 60 years' labor betwe 
Emmons and de Laguna. De Laguna has performed an invaluable service 
making Emmons's extensive Tlingit data, and his interpretations of them, availa 
for the first time. The Tlingit Indians is based on Emmons's decades of fieldwc 
in southeast Alaska and intimate knowledge of the Tlingit. As its publicatior 
more than a century after the beginning of his fieldwork, the work has an ar- 
chival quality and represents a valuable time depth in the understanding of the 
Tlingit. This quality is enhanced by the copious quotations from early sources 
regarding Tlingit life in the eighteenth and early nineteenth centuries, including 
Russian sources, and other observations of the Tlingit in the late nineteenth cen 



Numerous 



il- 



many sketches by Emmons himself 



dress, housing, transportation, and technology. Archival photos also illustrate 



shamanistic curing, ceremonial costume 
totem poles. The volume 



rial 



Tlingit 



The Tlingit Indians is a comprehensive ethnography. Chapters cover the land 
and people, social organization, villages, houses, forts, and other works, travel 



women 



es, aress ana decoration, the lite cycle, ceiem— - 
medicine, shamanism, witchcraft, games and game 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 277 



ing, and time 



climatic 



ceremonies, and medical practice. A bibliogr 



modern 



Emmons 



comments 



some 



amassed adds immensely to the value of the work, however. The segregation 
of the tables at the back of the volume is a minor inconvenience. 

Valuable for the ethnobiologist are tables of edible marine resources (in- 
vertebrates and algae), edible plants, colors, paints, dyes and stains, basketry 
materials, and medicines. De Laguna has enlisted the services of the ethnobotanist 
Alix Wennekens to supply scientific names for the plants discussed by Emmons, 
which is very helpful. The addition of scientific names in the text for plant foods 
and medicines, as well as in the tables, would enhance the ease of use of the 
book for ethnobiologists. The discussions of harvesting and processing food plants 
and basketry materials are extremely useful. Good treatments of traditional hun- 
ting and fishing techniques are also included. A minor error in the food plant 
section is the identification of the edible fern root as lady fern (Athyrium filix- 
femina). Turner et al. (1992) have demonstrated that the fern rootstock collected 
by the Tlingit is Dryopteris expansa. 

The Tlingit Indians is a good value for anyone interested in the cultures of the 
Northwest Coast of North America, or in hunter-gatherer subsistence and 
technology. The book is also valuable for those with an interest in traditional 
medicines, as it has extensive sections on diagnosis and treatment of illness, 
shamanism, and witchcraft. 



LITERATURE CITED 



TURNER 



CESKA 



ethnobotanical puzzle. Journal of Ethnobiology 12:1-34. 



Leslie M. Johnson Gottesfeld 
Department of Anthropology 
University of Alberta 
Edmonton, Alberta, Canada 



278 



BOOK REVIEW Vol. 12, No. 2 



BOOK REVIEW 



The Seven Sisters of Sleep. Mordecai C. Cooke. Lincoln, MA: Quarterman 
Publications, Inc. (P.O. Box 156, Lincoln, MA 01773), 1989 (reproduction of 
original 1860 edition). Pp. xi, 371. $45.00 (clothbound). ISBN 0-88000-146-1. 

Mordecai Cubitt Cooke's first book, The Seven Sisters of Sleep, originally 
published in 1860, is one of the rarest classics of the psychoactive drug literature, 
with only a few copies known to exist in U.S. libraries (Cooke 1860). The third 
book to survey the topic and the second in English, published just five years after 



»/ Common Life (Johnston 1855) and Die Narkotischen 
Mensch (Bibra 1855), The Seven Sisters of 



simile 



Michael 



The book lacks a bibliography or index. There is a frontispiece photograph 



man 



Quarterman is to be commended 



this rare book, which will be followed by an English translation of von Bibra's 



1855 classic. 

The 

discussed in the book 
hemp or marij 



opium (Papaver somnifi 



spp.j, thornapple (Datura spp.) and "the exile of Siberia," the fly-agaric (Amanita 
muscaria L. ex Fr.) (Pers. ex Gray). In his opening chapter, Cooke introduces his 



charming legend 



he invented 



some n 
making 



history 



and ethnobotany of tobacco, Cooke discourses on "pipeology," the history of 
tobacco pipes, then devotes individual chapters to tobacco snuffing (in which he 
mentions niopo, entheogenic Anadenanthera snuff from the Orinoco) and tobacco 
chewing. A brief chapter on tobacco substitutes leads into six chapters on opium 



history, ethnopharm 



Revels 



Pandemonium"). There 



m Morals" and a chapter on opium 



history, effects, and ethnopharmacologv, followed by three chapters 



similar information 



), "Datura 

and there 



chapt 



alimentary 



chapter, and tr 
harv and medic 



use of "Whitewash and Clay," the second on medicinal use of "Precious Metals 
in this case the not-so- precious arsenic and mercury. A final chapter, in which 



crumbs 



for 



the tobacco habit, and the fabled angel of sleep tenderly embraces 
death as the curtain closes; to an encore Appendix of 19 tables. 



gel of 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 279 



This quaint and witty book is well written and provides an entertaining 
introduction to the subject for the nonspecialist. Cooke's florid and whimsical 
style will delight the lover of Victorian literature. The book presents solid, though 
dated information on the "seven sisters," and mentions many other, lesser- 
known inebriating plants. Although this wonderful book has been a lost classic 
for the past century, it has had a lasting impact on our culture. The Seven Sisters 
of Sleep is widely regarded to have inspired Lewis Carroll in the writing of Alice's 
Adventures in Wonderland —especially the scene of the hookah-smoking caterpillar 
sitting on a mushroom which could make Alice grow larger or smaller, just as 
Cooke's "Exile of Siberia" was said to give "erroneous impressions of size and 



distance" 



ing in the road becomes a formidable 



a leap is taken sufficient to clear a barrel of ale ... 



LITERATURE CITED 



BIBRA, ERNST FREIHERRN VON. 1855. Die Narkotischen Genussmittel und der Mensch. 

Verlag von Wilhelm Schmid, Nurenberg. 
COOKE, M.C. 1860. The Seven Sisters of Sleep. Popular History of the Seven Prevailing 

Narcotics of the World. James Blackwood, London. 
JOHNSTON, JAMES F. 1855. The Chemistry of Common Life. D. Appleton and Co., 



New York. 



Jonathan Ott 
Natural Products Co. 
Apartado Postal 274 
Xalapa, Veracruz, Mexico 



BOOK REVIEW 



Medicinal Wild Plants of the Prairie: An Ethnobotanical 

schpr I awrpnrp- fTnivcreifv Prt»ss nf Kansas (2501 W. 



maps, 43 line drawings by Willia 



Whitney. $25.00 (clothbound) ISBN 0-7006-0526-6. $9 
0-7006-0527-4. 



companion 



plants of the North American plains (Kindscher 1987), presents 43 monograpns 
on important medicinal species. Each monograph averages five pages and features 
a full-page line drawing of the plant, a small distribution map, and the following 
sections: Common Names. Indian Names, Scientific Names (including exp ana- 
Habitat, Parts Used, 



name and common synonyms), Description 
Anglo Folk Use, Medical History, Scientifu 



280 



BOOK REVIEW Vol. 12, No. 2 



vation. In general, the descriptions (of a single paragraph) together with the 
illustrations are adequate for identification purposes. Accounts of Indian and 
Anglo folk use are fairly comprehensive, as are the notes on medical history. 
Information on cultivation is rather sketchy in most cases, but is at times based 
on the author's personal experience in Kansas. A six-page glossary will aid the 
amateur with botanical and (at times quaint) medical terminology, and all sec- 
tions make reference to a bibliography (replete with full journal and article titles, 
plus page numbers) of 259 sources, which covers the major publications on North 
American medical ethnobotany. The book ends with a useful index of 26 pages. 
AftPr thp first sprtion of monographs of the most important "Medicinal Wild 



Medicinal 



mini-m 



map 



name of the plant is given, followed by the name 



family and its popular name. A brief description is followed by a parag 
two detailing ethnobotanical use. Data on medical history, phytochemistry 
cultivation round out each mini-monograph. 



The book is handsomely 



sm 



presum 



sm 



book is more than justifiable. As it is, I would recommend the $25.00 hardcover 
over the $9.95 paperback for reasons of durability. 

I recommend this book heartily to all ethnobiologists interested in ethno- 
medicine. Phytochemists will find the book useful, but by no means complete 
and up-to-date with regard to chemical information on these plants. Physicians 
and others attracted to herbal medicine will find much of interest here, but would 
do well to heed the frontispiece disclaimer which advises that "this book pro- 
vides only limited information" on medicinal use and "contains descriptions, 
not prescriptions." Information on the use of some of these plants or their relatives 
as inebriants seems to have been ignored or glossed over. For example, Cree use 
of Acorus calamus L. as a stimulant and entheoeen ("hallucinogen") is not men- 



tioned (there is a reference to constituent teta-asarone as "a mild hallucinogen,' 
although this benzene derivative is not found in North American strains usee 
as inebriants). Traditional uses of Artemisia ludoviciana Nutt. suggestive of psycho 



much 



American 



i flat a 



emphasized. Similarly, the Hopi "hallucinogen" Mirabilis multiflot 
is not mentioned in the section on M. nyctaginea (Michx.) MacM. References to 
the use of these plants as inebriants can be found in a recent survey of North 
American medicinal plants (Moerman 1986). Finally, the author neglects to men- 
tion the fact that Desmanthus illinoensis (Michx.) MacM. and Lespedeza capitata Michx. 
contain the entheogenic drug N,N, -dimethyltryptamine (DMT). Since DMT is 



Winter 1992 JOURNAL OF ETHNOBIOLOGY 281 



illegal, so, technically, are these plants (along with some 300 other species now 
known to contain scheduled drugs), which might influence the reader consider- 
ing growing these species in a home garden. But these are minor points, and 
in no way detract from a useful and valuable book, which will be a significant 
additon to any ethnobiologist's library. 



LITERATURE CITED 



KINDSCHER, KELLY. 1987. Edible Wild Plants of the Prairie: An Ethnobotanical Guide. 

University Press of Kansas, Lawrence, KS. 
MOERMAN, DANIEL E. 1986. Medicinal Plants of Native America. Research Reports 

in Ethnobotany, Contribution 2; Technical Reports No. 19. Museum of Anthropology 



University of Michigan, Ann Arbor, MI. 



Jonathan Ott 
Natural Products Co. 
Apartado Postal 274 
Xalapa, Veracruz, Mexico 



282 



Vol. 12, No. 2 



NOTICE TO AUTHORS 



Journal of Ethnobiology has published "Guidelines for Authors" in 



Volume 10, Number 2 (Winter 1990). Many 



the Journal 



the Journal 



write 



must submit two cooies of their manusc 



submitted 



Submit m 



DEBORAH M. PEARSALL, Editor 

Journal of Ethnobiology 
American Archaeology Division 

103 Swallow Hall 

University of Missouri 

Columbia, Missouri 65211 USA 

FAX: 314-882-9410 



Submit manuscripts written in the Spanish language to: 

SR. ALEJANDRO DE AVILA B, Associate Editor 

Journal of Eth nobiology 
Centro de Graduados e Investigacion 

A.P. 1378 
68000 Oaxaca, Oaxaca, Mexico 



NEWS AND COMMENTS 

Individuals with inform 
the Journal should submit all 



and Comments" section of 
to Gary J. Martin, 94, Blvd. 



FAX: 33-1-44919882 



REVIEWS 



We welcome suggestions on books to review or actual 
ship of the Journal. Please send suggestions, comments, 
Journal's book review editors. Please see inside fron 
addresses. 

SUBSCRIPTIONS 



Subsc 



iptions to the Journal of Ethnobiology 
Department of Anthropology, Univ 



•sity of Nevada, Reno, NV 



option rates are $60.00, institutional; $25.00 individual suDscnDers rrum 
America; $25.00 students subscribers; $35.00 regular individual subscribers 
cept for Latin America; Joint member (spouse; one copy of journal), add $10.00; 
stage: $8.00 (outside of U.S.A., Canada, and Mexico). Write checks payable 
Journal of Ethnobiology. Defective copies or copies lost in shipment will be 



copies or copies 
within 



CONTENTS 

EDITOR'S VIEW * 

ETHNOECOLOGY, BIODIVERSITY, AND 
MODERNIZATION IN ANDEAN POTATO AGRICULTURE 

Stephen B. Brush 161 

THE USE OF SOUND RECORDINGS AS 
VOUCHER SPECIMENS AND STIMULUS 
MATERIALS IN ETHNOZOOLOGICAL RESEARCH 

Eugene Hunn 1°' 

PREHISTORIC MEDICINAL PLANT USAGE: 
A CASE STUDY FROM COPROLITES 

Kristin D. Sobolik and Deborah }. Gerick 203 

RECENT DOCTORAL DISSERTATIONS OF INTEREST 
TO ETHNOBIOLOGISTS: FALL 1991-FALL 1992 

Terence E. Hays and Joseph E. LaFerriere 213 

SHORT COMMUNICATION 

Gary J. Martin and Sergio Madrid 227 

ABSTRACTS OF PRESENTATIONS 235 

NEWS and COMMENTS 273 

BOOK REVIEWS 185, 198, 202, 211, 224, 225, 232, 234, 

271, 275, 276, 278, 279 



V 



V^\\ ^ 



J5°it 







VOLUME 13, NUMBER 1 



SUMMER 1993 



Journal and Society Organization 

EDITOR: Deborah M. Pearsall, American Archaeology Division, 107 Swallow Hall, Univer- 
sity of Missouri, Columbia, MO 65211. 
ASSOCIATE EDITOR (Spanish): Alejandro de Avila B., Department of Anthropology, 

University of California, Berkeley, CA 94720. 
NEWS & COMMENTS EDITOR: Gary J. Martin, 94 Blvd. Flandrin, 75116, Paris, France. 

FAX: 33/1/45533001. 
BOOK REVIEW EDITOR: Carlos E.A. Coimbra, Jr., Escola Nacional de Saude Publica- 

FIOCRUZ, Fundacao Oswaldo Cruz, Nucleo de Doencas Endemicas, Rua Leopoldo 

Bulhoes-Manguinhos, 21.041 Rio de Janiero-RJ-BRASIL. 
BOOK REVIEW EDITOR: Nancy J. Turner, Environmental Studies Program, P.O. Box 1700, 

University of Victoria, Victoria, B.C. CANADA V8W 2Y2. 
PRESIDENT: Cecil H. Brown, Department of Anthropology, Northern Illinois University, 

DeKalb, Illinois 60115. 
PRESIDENT-ELECT: Catherine S. Fowler, Department of Anthropology, University of 

Nevada, Reno, Nevada 89557. 

SECRETARY/TREASURER: Brien A. Meilleur, Amy B.H. Greenwell Ethnobotanical Gar- 
den, Bishop Museum, P.O. Box 1053, Captain Cook, HI 96704. 

CONFERENCE COORDINATOR: Jan Timbrook, Department of Anthropology Santa 
Barbara Museum of Natural History, 2559 Puesta Del Sol Road, Santa Barbara, CA 93105. 

BOARD OF TRUSTEES 

ROBERT A. BYE, JR., Universidad Nacional Autonoma de Mexico, MEXICO: ethnobotany 

eth noecology. 
TIMOTHY JOHNS, Macdonald College of McGill University, CANADA. 
Ex officio: Past Presidents Steven A. Weber, Amadeo M. Rea, Elizabeth S. Wing, and Paul 

Minnis; Permanent board member Steven D. Emslie; The Editor, President, President 

Elect, Secretary/Treasurer, and Conference Coordinator. 



EDITORIAL BOARD 



ADAMS 



EUGENE N. ANDERSON, University of California, Riverside, USA; ethnobotany. 



cal ethnobotany. 
VID R. HARRIS, I 

terns, archaeobotany. 



ethnobiological classification 



TIMOTHY 



HARRIET V. KUHNLEIN, McGill University, CANADA; ethnonutrition, human nutrition. 
GARY J. MARTIN, Grupo de Apoyo al Desarrollo Etnico, Oaxaca, MEXICO; ethnobiologic 



ification 



Museum 



sity ENGLAND; natural resource management, ethnoecology, ethnoentomology, tropical cul- 



tural ecology. 



ology 



San Diego Natural History Museum, USA; cultural ecology, 



MOLLIE S 
WILLARD 



J 



Feature editors Carlos E.A. Coimbra and Nancy J. Turner (see above). 

Journal of Ethnobiology is published semi-annually. Manuscripts for publication, information for the "News an 

CnmmpnK" and hn/-»L- rowiom e^i^*^ ^V».-*.,M U« ™„». ».~ *.u~ ««-^-^«.-;^».„ ^j;*.^, n ~ tU** IndAo Viark rover of thlS ISSUc. 



®Society of Ethnobiology 
* ISSN 0278-0771 



Journal of 
Ethnobiology 



MISSOURI BOTANICAL 



NOV 08 1993 



GARDEN UBRARX 



VOLUME 13, NUMBER 1 



SUMMER 1993 



CONTENTS 



EDITOR'S VIEW 



MIDDEN AND COPROLITE DERIVED SUBSISTENCE 
EVIDENCE: AN ANALYSIS OF DATA FROM THE 
LA QUINTA SITE, SALTON BASIN, CALIFORNIA 

MarkQ. Sutton 



NEW PERSPECTIVES ON A WILD GOURD 
IN EASTERN NORTH AMERICA 

C. Wesley Cowan and Bruce D. Smith 



ISOZYMIC CHARACTERIZATION OF WILD 
POPULATIONS OF Cucurbita pepo 

Deena S. Decker-Walters, Terrence W. Walters, 

C. Wesley Cowan, Bruce D. Smith 



HISTORY AND GEOGRAPHIC DISTRIBUTION OF 
Cucurbita pepo GOURDS IN FLORIDA 

Lee A. Newsom, S. David Webb, James S. Dunbar 



1 



1 



17 



55 



75 



NEWS AND COMMENTS " 



BOOK REVIEWS 15, 54, 73, 97, 131-147 



Announcement 



SOCIETY OF ETHNOBIOLOGY 
SEVENTEENTH ANNUAL CONFERENCE 

March 16-18. 1994 



Victoria, British Columbia, CANADA 



Sponsored by: 
Royal British Columbia Museum 

and 
Environmental Studies Program 

University of Victoria 



A major theme will be : 
Sustainable Land Management and Harvesting 

Methods of Indigenous Peoples 



further information, please conta 
Nancy J. Turner 

Environmental Studies Program 
P.O. Box 1700 

University of Victoria 

Victoria, British Columbia 

CANADA V8W 2Y2 

phone: (604) 721-6124 

FAX: (604) 721-8653 

E-MAIL: nj turner @sol.UVic.C A 




With this issue of the Journal we initiate a new feature in the "News and 
Comments" section, "Opinions." As announced by News and Comments editor 
Gary Martin in the last issue, we open an editorial space for readers to submit 
short (less than 500 words) opinion pieces on issues of concern to ethnobiologists. 
Gary will select and edit this section; opinion pieces will not be sent for outside 
review. "News and Comments" is quite full this issue, and includes a stimulating 
exchange between Glenna Dean and Karl Reinhard on the use of pollen con- 
centrations in coprolite analysis. Readers are encouraged to comment on articles 
published in the Journal; such comments are subject to peer review and response 
by the authors of the original articles. 

I am pleased that three articles focused on wild Cucurbita gourds of eastern 
North America, generated from papers presented at the 1992 Society of Eth- 
nobiology meetings in Washington, D.C., all appear in this issue. With apologies 
to those readers who may tire of the wild gourd, publishing these papers together 
presented a rare opportunity for the authors, and for the Journal, to showcase the 
application of multiple lines of evidence to an ethnobiological research problem. 

I would like to acknowledge the hard work of Janis Alcorn and Cecil Brown, 
who are leaving the editorial board, and to welcome Gene Anderson to the board. 
The review process only works through the joint efforts of board members and 
reviewers; I thank everyone for their efforts and invite any reader who would like 
to be part of the reviewer pool to contact me. 

Finally special thanks to Linda Desmond, friend of the Society, former owner 
of I'm Your Type and Graphics. Before I became Journal editor, I never thought 
about typesetters, the skilled people who take marked up manuscripts, full of 
inserts, arrows, and words in exotic languages, and turn them into pages ready 
for printing. Linda typeset the Journal from practically its beginning until this 
issue. She made the editor's job easier; she made us look good. Thanks, Linda. 



DMP 



/. EthnobioL 13(1): 1-15 



Summer 1993 



MIDDEN AND COPROLITE DERIVED SUBSISTENCE 
EVIDENCE: AN ANALYSIS OF DATA FROM THE LA QUINTA 

SITE, SALTON BASIN, CALIFORNIA 



MARK Q. SUTTON 

Department of Sociology and Anthropology 
California State University, Bakersfield 

9001 Stockdale Highway 
Bakersfield, CA 93311-1099 

ABSTRACT.— An analysis of resource constituent remains recovered in coprolites 
from sites in the Salton Basin, California, reveals several patterns of food availabil- 
ity, preference, and utilization. Specific combinations of foods are noted, indicat- 
ing possible "meals." The coprolite data, combined with traditional faunal and 
floral analyses, form a more comprehensive view of subsistence. It is recom- 
mended that noncoprolite data be integrated into coprolite studies. 

RESUMEN.— Un analisis de los restos de los materiales constitutivos recuperados 
en coprolitos encontrados en algunos sitios de la cuenca del lago Salton, en 
California, muestran varios patrones de disponibilidad, preferencia, y utilizacion 
de alimentos. Se notan combinaciones especificas de alimentos, indicando posi- 
bles "comidas." Los datos derivados de los coprolitos, combinados con los analisis 
tradicionales de fauna y de flora, proporcionan una vision mas completa de la 
subsistencia. Se recomienda que a los estudios sobre coprolitos se integren datos 
no derivados de los coprolitos mismos. 

RESUME.— Une analyse en resources constituantes des excrements humains, 
recuperes en coprolites des sites dans le Bassin "Salton" en Californie, revele de 
differents modeles de la disponibilite des aliments preferes et utilises. Des com- 
binaisons specifiques des aliments sont constatees indiquant des "repas" pos- 
sibles. Les donnees sur le coprolite, combinees avec des analyses traditionnelles 
de la faune et de la flore, constituent une vue de subsistence plus complete. C'est a 
conseiller que les donnees noncoprolites soient integrees dans les etudes coprolites. 



INTRODUCTION 



human fecal matter 



information regarding prehistoric diet, nutrition, health, and pharmacology (see 
Fry 1985, Sobolik 1990, and Reinhard and Bryant 1992 for recent reviews of cop- 
rolite studies). Unfortunately, coprolites are very fragile and susceptible to decom- 
position, and so rarely are recovered archaeologically. 

Coprolites form direct evidence of substances consumed, although not always 
as food, as opposed to standard faunal and floral remains, which form indirect 
dietary evidence. Archaeologists studying coprolites make a number of assump- 
tions, often with great merit, regarding the nature and origin of the specimens. 



2 



SUTTON Vol. 13, No. 1 



materials 



came and that such materials 



tied. Secondly, coprolites usually are viewed as largely representing the subsis- 



ceremonial 



medicinal purposes beine more 



Third, it is assumed that each specimen represents a unique elimination event 
and is not mixed or combined with other such events. In spite of this, obvious 
fragments, possibly representing separate events, frequently are grouped together 
as one specimen for analysis. Further, it generally is assumed that materials pres- 
ent in a coprolite represent the food consumed within the 24-hour period preced- 
ing its deposition (e.g., Fry 1985:128), although this may not be the case (e.g., 
Sobolik 1988a:207; Jones 1986). As such, coprolites likely are a combination of 
several meals (e.g., Watson 1974:240). 

Other factors are of note in coprolite analysis (see Sobolik 1988b: 114). As the 
surviving (i.e., visible) materials are those that were not digested, only the indi- 
gestible part of the diet is visually represented and we do not understand all the 
taphonomic problems (i.e., digestion, processing, preservation, and so on) associ- 
ated with coprolites. For example, large mammals will not be visually represented 
in coprolites, nor will animals that have been subjected to certain types of process- 
ing ( e -g- filleting fish). However, this situation is changing with the addition of 
the immunological technique that can identify nonvisible constituents (Newman 
et al. 1993). Coprolites may be discovered singly or in concentrations that proba- 



While 



homogeneous, this may 



customs 



sample would be skewed and the interpretations 



may 



's appear to assume sam 
from Nubian mummies i 



exam 



Most researchers focus on the inter-specimen 



sample of coprolites. Relative abundance is assumed 



lm 



combination and utilization (i.e., intra-specimen 



The goal of the present study is to determine 

combinations within a sample of coprolites from the La Quinta site (CA-I 
in the Salton Basin, California and to integrate noncoprolite (i.e., midden 
faunal and floral data into an overall view of site-specific subsistence. 



THE SALTON BASIN DATA BASE 



Coprolites have been recovered from six open sites in the northern Coachella 
Valley as part of excavation projects (Fig. 1). All sites lie within the ethnographic 
territory of the Cahuilla Indians (Bean 1978) who probably occupied the region at 

least Sincp thp final clanA n( T »1«, r-^u..:n_ t?nn /unii.. irvro\ TUo 



some 500 years ago (Wilke 



analysis of each of these coprolite series was conducted by first rehydrating the 
specimens in a solution of trisodium phosphate. Specimens then were filtered, 
dried, and passed through a series of small screens. Recognizable constituents 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



3 






1 1 



*•%> 



<t^ CA-RIV-3793 
G> V ^^- Myoma Dunes 



i 



CA-fi/V-3682 



I 



CA-RIV-1179 



Wadi Beadmaker 



>\i\\ 



CA-RIV-2827 



» I 



// 



■ « « 



• 9 



• •■••••• 









II 



• • 



I 






• ■ 



I 



i ■ 



• • 



• • • 



• • 



LOCATION OF SITES 
DISCUSSED IN TEXT 



#WA 



</ 






/ 



/ 



M 






••••••••••• 



if 



/\nza Borrego 
Desert 



\\fy 



V\ I // 



^% 



• • • 



^ \ 

v ^ ,. >^ •;;: 

iV I • 

ji\\\ / • 

m \ r. *•• 

\ x / mi 

/ • ! 



# • • • •*• ••••••• 



• ••■••»•■"----- 
....... •••••••• 



Lake 



from Wilke 



were sorted and identified. Relative abundance was estimated following estab- 
lished techniques used in wildlife biology (see Wilke 1978:154-157 for a complete 
description of analytical techniques). 

Three of the sites, CA-RIV-3682 (Yohe 1990), CA-RIV-3793 (Goodman and 
Arkush 1990; Goodman 1990), and CA-RIV-2827 (Sutton and Wilke 1988a; Farrell 
1988) are small and contained limited assemblages of artifacts, ecofacts, and 
coprolites. The other three sites contained much larger numbers of coprolites plus 



other faunal and floral data. 



Myoma 



located in mesquite-anchored sand dunes along tne nonnenuiiuai MH »« ~. — . ~ 
Cahuilla and generally dates to the final stand of the lake, approximately 



Many 



The site is located on the valley floor and is not directly adjacent to upland hao- 



materials 



(Wilke 



4 



SUTTON Vol. 13, No. 1 



The second large site, Wadi Beadmaker, is the remnant of an extensive camp 
located along the northeastern shore of the lake; it also dates to the final lake- 
stand. Excavation at the site resulted in recovery of numerous artifacts, ecofacts, 
and approximately 70 coprolites. As with Myoma Dunes, analysis of materials 
recovered from the site was limited to the coprolites (Wilke 1978) and no comple- 
mentary ecof actual data were reported. 

The third site, the La Quinta site (CA-RIV-1179), is located in an ecotone of at 
least three environmental zones (lake shore, desert, and mountain) along the north- 
western shore of the former lake. The site was excavated in 1985. La Quinta con- 
sisted of a fairly large open camp dating from the final stand of Lake Cahuilla (ca. 
A.D. 1500) and contained numerous artifacts, ecofacts, cremations, and 128 copro- 
lites. A full analytical report on the recovered materials was produced (Sutton and 
Wilke 1988a); this is the only such comprehensive report for a major site in the region. 

Farrell (1988) analyzed 30 coprolites from the La Quinta site. Most were dis- 
covered in a relatively small area, suggesting the presence of a latrine. Macro- 
scopic floral and faunal elements were identified to taxon where possible, the 
remainder being classified as unidentified fragments (Farrell 1988:132-133). Sev- 
eral specimens appeared to consist primarily of pollen, which was identified; 
however, no general pollen or phytolith studies were conducted on the samples. 
Abundance of materials recovered from the coprolites was ranked as abundant, 
frequent, infrequent, or trace based on the volume of material in each specimen. 

Farrell (1988) noted fish bone in all analyzed coprolites. Two species, bony tail 
chub (Gila elegans) and razorback sucker (Xyrauchen texanus), were identified. Two 
other fish, the Colorado squawfish (Ptychocheilus lucius) and mullet (Mugil cephalus), 
also were present in prehistoric Lake Cahuilla. Mullet remains are abundant at some 
other lakeshore sites (Follett 1988:154) but were not identified at La Quinta. Squaw- 
fish remains were observed in the midden at the La Quinta site (Follett 1988:154). 

Seven examples of articulated fish vertebrae were recovered from the La Quinta 
midden (Follett 1988); six bonytail chub and one razorback sucker. Five of the six 
chub examples consist of caudal vertebrae, indicating that tails had been removed 
and discarded (unconsumed?). The sixth chub specimen consisted of eight pre- 
caudal vertebrae. The razorback sucker specimen consisted of (apparently) pre- 
caudal vertebrae. This could possibly be the remains of a filleted fish. 



THE CURRENT STUDY 



Ihe objective of this study was to conduct a comparati 
Quinta coprolite constituents (from Farrell 1988) to determine 



might include food coml 
e or habits (i.e., meals) 



that could be used to delineate dietary preference or habits (i.e., meals) and dif- 
ferences in the seasonal use of resources. Faunal and floral materials recovered from 



^ — — - — ~ ^** *»^*v. ni^n ^v^upaicu ikj 11 ic LUU1 

discover additional patterns between the two data sets. 



attempt 



Methods 



Q 



1988) were compared using a hierarchical cluster analysis, part of SPSS-PC 



Computer). Mem 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



5 



TABLE 1.— Coprolite clusters by constituent, CA-RIV-1179. 



Specimen 



Cluster One 

6 (24-32) 

22 (16-48) 

12 (19-33) 

7 (23-33) 

10 (19-38) 

3 (25-14) 

29 (SC-4b) 
2 (25-22) 
21 (16-49) 

30 (SC-4a) 

9 (23-30) 

27 (16-32) 

14 (17-51) 

Cluster Two 

5 (24-35) 
20 (16-67) 

4 (24-36) 
1 (25-26) 

17 (17-45) 

18 (17-26) 

28 (4-6) 

Cluster Three 

11 (19-35) 

19 (16-68) 
16 (17-46) 

23 (16-43) 

15 (17-47) 
26 (16-34) 

13 (19-24) 
25 (16-36) 

Cluster Four 

8 (23-32) 

24 (16-39) 



Resource 23 
ABCDE FGH I J KLMNOPQRS 



0010000000004000001 
0010000000004000001 
1010000000004000001 
1010000000014000001 
1010000000014000001 
0020000000004100001 
0010000000004100001 
0020000000004000001 
0010000000004100011 
101000 0000004100011 
0030000000014000011 
0030000001003100001 
0040000100014110113 



0030000000000100001 
0030000000100000001 
0040000000000000001 
0040000000000100012 

0020001000000000001 
0031000010010100004 

0040100000011000014 



4040000000010000002 
4040000000010100002 
4040000000000100012 
4040000000001210012 
4041000010002100003 
4041000010002100003 
3030000000003100012 
4040010000004001014 



0440000000000000013 
0440000000104100002 



Computer specimen numbers (1-30; 
2 Taxa list: 

A bony tail (Gila elegans) 




dendrogram); catalog numbers in parentheses 



H unident. mammal 



O dicoria (Dicoria canescens) 



B razorback (Xyrauchen texanus) I unident. vertebrate P mesquite (Ptosopis spp.) 

C unident. fish 1 mussel shell (Anodonta) Q goosefoot (Chenopodmm) 

D tortoise (Xewbates agassizii) 
E chuckwalla (Sauromalus obesus) 
F unident. reptile 
G cottontail (Sylvilagus audubonii) 
3 Abundance codes: 4 = abundant (A); 



J mussel shell (Anodonta) 
K land snail (Physa) 
L unident. insect 
M cattail (Typha) 
N bulrush (Scirpus) 



R unident. seeds 
S charcoal 



infrequent (I); 1 = trace 



6 



SUTTON Vol. 13, No. 1 



cluster was based on nearest neighbor to the center of the clusters, defined as 



Numeric 



abundance rankings (abundant, frequent, infrequent, trace) given in the original 
study, with zero used to designate absence. 



Results. 



main 



markably 



Cluster One. Cluster One (n = 13) is dominated 



comprising 



specimen) in 12 of 13 sam 



fish, all identified as bonytail chub, are present; bulrush seeds (or tule; Scirpus) are 
present in trace amounts in six of the samples. No reptile remains, and onlv one 



amounts 



mammal, are present in these specime 



maj 



month 



; the primary constitueni 
some pollen may remain 



amounts may be residuals from earlier meals 



lm 



may 



animal resources appear not to have been consumed 



consumed 



able fresh from May to July (Shreve and Wiggins 1964:229), suggesting that the 
Cluster One coprolites date from that season. Fish also should have been available 
in quantity during that time and their relative paucity may be the result of people 
concentrating on the collection of cattail. 

Cluster Two. This cluster (n = 7) is dominated by unidentified fish which are 
mostly charred. No elements could be identified to genus. Present also are the 
only cottontail (Sylvilagus) and chuckwalla (Sauromalus; three scapulae in one 
specimen) remains identified during the study. Few plant resources were identi- 
fied in this cluster and charcoal is abundant in only two specimens. 

The specimens comprising Cluster Two may reflect a diet centered on the con- 



small terrestrial animal 



(albeit few) terrestrial remains. 



remains 



The 



suggests that fish had been processed; perhaps fillets were made 



infreq 



~~..^ „..« .i U i C4 u C m Liuircoai oemg me result ot the drying process) or tisn (ana 
bones) were ground on a metate. If this interpretation is correct it suggests the 

COnSUmnHcm nf Qtrn-orl (; c u 



consumption of stored fish. 

1 ■ ■ m 



In light of the possibility discussed below, that bonytail were processed in two 
?rent ways, one resulting in the elimination of mnct k™ q ;» ia ^occihlo that the 



most bone, it is possible 



lm 



Most visible remains (bones) were simply absent. Protein (immu 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



7 







Rescaled Distance Cluster Combine 



5 



10 



15 



20 



25 



+ 



Sample No 

6 
22 
12 

7 
10 

3 
29 

2 
21 
30 

9 
27 
14 

5 
20 

4 

1 
17 
18 
28 
11 
19 
16 
23 
15 
26 
13 
25 

8 
24 



n 



-J h- 










1 


J 

i 


Cluster One 




1 








i 

i 






































1 
















































/"•I 1 1 « ♦" £» "V TV.Trf-N 








¥ v^^k m^^^^ 
















fl iipfar T , V»T*Qa 






v^ J. Lis 
















1 











































Cluster Four 



FIG. 2.— Dendrogram illustrating the clustering of coprolite constituents from the 
La Quinta site (CA-RIV-1179). 

nological response) studies on coprolite matrix may be useful for addressing this 
possibility (e.g., Newman et al. 1993). 

If one were to view the fish remains as evidence for consumption of stored 
foods, a late winter/early spring season of deposition may be indicated. The pres- 
ence of chuckwalla remains suggests spring or later (Wallace 1978:109). 

Cluster Three. Cluster Three (n = 8) is dominated by bonytail and unidentified 
fish remains (mostly charred); razorback sucker was not identified in the cluster. 
Tortoise and unidentified reptile are present, as is unidentified vertebrate bone. 
Cattail and bulrush seeds are often present, but only once in quantities considered 
abundant. Charcoal is present above trace amounts in each of the specimens. 



sam 



consumed 



8 



SUTTON Vol. 13, No. 1 



many bones were charred (Farrell 1988:137). This suggests that fish were placed in 
an open fire to cook and consumed partially charred. 

Bonvtail remains from the midden, however, are mostlv uncharred. sueeest- 



(Wilke 



meat 



consumed; or (2) baked with meat beine remove 



eaten and uncharred bones discarded. 



small 



samples and may have been consumed 



bonytail. Bulrush produces seeds between May and August (Munz 1974:902), 
overlapping occurrence with cattail, although Farrell (1988:135) thought that 



some coprolites from Myoma Dunes ( Wilke 



formed a maj 



Four. The 



fourth cluster (n = 2) contains abundant razorback and un- 
identified fish remains, bonytail not being identified in either specimen . Charcoal also is 
present in greater than trace amounts . Cattail seed is abundant in one of the specimens . 
The general absence of razorback suckers in the coprolites is interesting since 
they are much more common in the general midden (Follett 1988). Suckers contain a 
large number of small bones and may have been processed differently than bonytail 
(e.g., filleted and broiled instead of baked whole; see McGinnis 1984:294 for obser- 
vations in this regard). Thus, it is possible that razorback was a more important 
resource than indicated in the coprolites. 



^^ uoolU1 '- «*» icnwins were consistently present in all samples altnougn 
their abundance and condition (identification) varied considerably. Even if whole 
fish are consumed, most bone is digested (i.e., 90%; Jones 1986:55) and so not 



t_7 \ •/ ** 

elements of a coprolite. While 



values are unaffected 



assum 



uniform 



from 



importance of fish (and other aquatic) resources appears to have changed season- 
ally, in spite of the presumed constant availability of fish (seasonal availability, if 
any of specific fish is unknown). 

^ Several combinations of resources were noted, forming, perhaps, the remains 
of "meals." Cattail (pollen, either alone or with flower heads) appears to have been 
consumed largely alone. Terrestrial animals seem not to have been consumed in 
meals with cattail, although some fish (mostly unidentified) was included. In addi- 
tion, bulrush often was identified in specimens containing bonytail. 



commo 



Wilke 1978; Farrell 1988), appear to have formed 



seems 



consumed 



samples. 



same pattern existed at Myoma Dunes Bed A; Wilke 
rback were filleted, that may account for the absence o: 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



9 



TABLE 2.— Floral remains from the midden recovered by flotation, CA-RIV-1179 
(from Swope 1988: Table 22). 



Origin 



Cat. No. 



Species 



Hearth 1 



Hearth 2 



Hearth 3 



Hearth 4 



Hearth 5 



Hearth 6 
Hearth 7 
Hearth 8 



Hearth 9 



108-4-6A Chenopodium, Juncus, Oligomeris linifolia, Prosopis glandulosa 



108-4-33 



108-8-21 



108-8-29 
108-12-25 



var. torreyana, Scirpus acutus, Scirpus, Sesuvium 
verrucosum, unidentified 

Chenopodium, Scirpus acutus, Scirpus validus, Sesuvium 
verrucosum 



108-8-9A Chenopodium, Juncus, Oligomeris linifolia, Prosopis glandulosa 

var. torreyana, Scirpus acutus, Scirpus validus, Sesuvium 
verrucosum, unidentified 

108-17-56 Chenopodium, Juncus, Scirpus acutus, Scirpus validus, 

Sesuvium verrucosum, unidentified 

Amaranthus, Juncus, Prosopis glandulosa var. torreyana, Scirpus 
acutus, Scirpus validus, Sesuvium verrucosum, Typha 

Scirpus acutus, unidentified 

Scirpus 



108-12-27 Chenopodium, Scirpus acutus, Sesuvium verrucosum 
108-12-34 Scirpus acutus , unidentified 

108-14-33 Scirpus acutus 

Soil Sample 108-16-73 Chenopodium, Juncus, Scirpus acutus, Scirpus validus, 

Sesuvium verrucosum 

Soil Sample A 108-19-21 Chenopodium, Juncus, Scirpus acutus, Sesuvium verrucosum 



Razorback was not identified in the same coprolite as bony tail. This is some- 
what intriguing since razorback is the larger fish (McGinnis 1984:148, 166). The 
historic Indian tribes of the lower Colorado River considered razorback a primary 
food fish (Castetter and Bell 1951:219) and procured them using the bow and 
arrow, nets, hook and line, and basketry traps (Castetter and Bell 1951:220-222). It 
is possible that razorback and bonytail were taken at different times, places, and/ 

or with different methods. 

Cluster Two contained few identified remains but considerable unidentified 



may 



mi 



m 



Other dietary evidence. 



from the remainder of the La Quinta 



materials 



remains from 



coprolites. However, several plants were found in the excavation samples that 



(h 



Their absence in the coprolite sam 



10 



SUTTON Vol. 13, No. 1 



TABLE 3. 

CA-RIV-1179 (from Sutton and Yohe 1988: Table 19) 1 



from the midden 



Xerobates agassizii 2 2 1 

Diposaurus dorsalis — 1 — 

Sauromalus obesus — — 3 

Order 

Podicipediformes 1—1 

Pelecanus cf . 

erythrorhynchos — — 1 

Anatidae — 2 

Anas sp. — — 

Fulica americana 4 



Dipodomys sp. 2 — — — — 1 

Neotoma sp. 1 — — — i 



Microtus californicus — — — — \ 

unident. rodent 4 14 — 4—1 

Canis latrans 5 — _ 3 _ _ 



Ovis canadensis 3 3 4 



*A11 units, depth in cm. 

includes eight awls (all artiodactyl) 



0- 10- 20- 30- 40- 50- 60- 70- 80- Crema- 
Scientific Name 10 20 30 40 50 60 70 80 90 tion Totals 



5 
1 
3 



2 



1 
2 



1 - 1 



4 



unident. bird 22 12 18 2 — 2 — 1 — 1 58 

5 
9 



Sylvilagus audubonii 5 

Lepus californicus 4 4 1 

unident. lagomorph 31 31 10 7 1 — — 2 — — 82 

Perognathus sp. 1_ !______ _ 2 

3 
2 

1 

23 

8 



4 14 



unident. artiodactyl 3— i_____i_ 4 9 

unident. mammal 100 40 31 14 2 2 — — 



_ 189 

Totals 188 109 72 30 5 6-4 1 9 2 424 



presumed consumpti 



of several resources not identified in the coprolites, notably bighorn sheep and 

i_*"lT ii-.. « ~ m « 1 1 



remains were much more common in the mi 

removal process was 



involved in the preparation of razorback. Of interest is the sudden decrease in fish 
remains in the upper portion of the deposit, while the remains of other animals 
increase. Sutton and Yohe (1988:113) suggested that this drop in fish remains "might 
reflect the decreasing availability of fish in conjunction with the desiccation of the 
lake, ca. A.D. 1500. It is [in] this later period that lagomorphs (mostly unidentified to 
genus) and birds (particularly quail) become the most numerous." 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 11 



A MODEL OF DIET AND SITE USE 



Based on the above patterns and observations the following model of season- 
ality and diet at the La Quinta site is proposed. 



Spring. —The site was first occupied in the spring, the inhabitants having come from 
an unidentified winter camp (possibly another lakeshore site such as Myoma 
Dunes). Small terrestrial animals and some fish were exploited. In addition, it is 
possible that larger land animals (e.g., bighorn sheep) were exploited but are not 
reflected in the coprolite constituents. A possible game diversion site, apparently 
for bighorn sheep, is located nearby (Sutton and Wilke 1988b). Perhaps people came 
to the site to harvest cattail (pollen and/or flower heads) and utilized other resources 
until the cattail was ready. When cattail pollen did become available (late spring/ 
early summer), it was heavily exploited. Cattail pollen formed the bulk of the diet 
during that time with other resources, including fish, being of secondary importance. 



Summer, —Cattail would have been exhausted in early to mid-summer, although 
some was perhaps stored. At that time fish and waterfowl were utilized, fish 
(primarily bony tail) in large quantities. The paucity of razorback in the coprolites, 
compared to its relative abundance in the midden, suggests a processing dif- 
ference between razorback and bony tail. 

Although fish formed the bulk of the summer diet, other animals and various 
plants were exploited and consumed. Some of these resources had to be obtained 
at somewhat distant localities, perhaps by special purpose task groups. Being 
located in an ecotone, the La Quinta site would have offered a variety of localized 
resource opportunities, perhaps making such trips relatively infrequent. 

Fall/ Win ten— There is no evidence that the site was occupied during the fall or 
winter. Desert dicoria (Dicoria canesccns), a winter staple (Wilke 1978:85), is largely 
absent in the coprolites, as are other resources thought to have formed part of the 
fall and winter diet (e.g., pinyon and mesquite; Wilke 1978:87). 

Thus, the inhabitants of La Quinta likely moved to another residential base 
camp(s) for the fall and winter. The location of such camps is unknown but might 
be in the uplands and/or another lakeshore location. A winter occupation is indi- 
cated at Mvoma Dunes, for example (Wilke 1978). 



Discussion . —There are a number 

current (i.e., the "visible") data. For exam 



two different ways (baked versus broiled) it would result in a differential distribu- 



While 



some 



immunolo 



utilized to test for presence of other animal meats, such as deer or mountain sheep, 
as the bones of such animals would not likely be present in identifiable fra S^ n ^ 
a coprolite. With this eeneral problem in mind, six coprolites from the CA-RIV-3682 



immunolo 



ewman 



12 



SUTTON Vol. 13, No. 1 



LATE PERIOD SETTLEMENT AND SUBSISTENCE 

ALONG NORTHERN LAKE CAHUILLA 



Wilke (1978:103) proposed a changing settlement-subsistence model for the 
late prehistoric of the northern Lake Cahuila basin. The model was based pri- 
marily on coprolite data from Myoma Dunes and ethnographic analogy, there 
being few other data available. That situation is still largely true, except for the 
results from La Quinta. While the La Quinta dietary and seasonality data come 
only from one site, they suggest that some revisions in the Wilke model may be in 
order. 



Wilke 



ystem 



permanent villages along the lakeshore for exploiting aquatic resources coupled 



camps 



ttlement 



remained basically the same (permanent 

permanent 



economic focus shifted from aquatic resources to terrestrial resources, likely re- 
sulting in increasing utilization of the surrounding uplands and a population 
increase in those areas (Wilke 1978:113). 

Quinta site was interpreted as a camp and not a permanent lakeshore 



Myoma Dunes (Wilke 



materials 



Cahuilla" (Wilke i 
sources (cf. Weide 



em 



Quinta site does not fit into the settlement 



Wilke 



em 



may 



m 



environments 



m 



posed by Wilke (1978). Whatever the case, the pattern at La Quinta is different 
than that found at Myoma Dunes or that of the historic Cahuilla. 

I suggest that the La Quinta site is part of a transitional system; the following 
factors support this view. First, the site dates to the very end of the last lakestand 



em 



113) 



more important later in time 



vidence that the La Quinta site served as a base camp, rather t 
pose camp, since the distribution of bighorn sheep remains 5 
animals were butchered elsewhere and taken to the La Quinta 



112 



prominent at La Quinta 



em. At some point in time 



from La Quinta and other camps 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



13 



lower elevations to remain close to the retreating lake. The later aspects of the 
transitional system would be represented at those sites, none of which have been 
identified and investigated. 



CONCLUSIONS 



People living around Lake Cahuilla at the time of its final stand (ca. A.D. 1500) 



com 



ment. There is little doubt that people camped near the lake to exploit the resources 
present there (e.g., fish, cattail, and waterfowl). It has been commonly assumed 
that in those situations fish was the dominant faunal food resource and that other 
animals were of secondary importance. 

Several interesting observations can be made from the analysis of coprolite 
and other dietary data. First, fish were apparently not a primary resource at all 
times while people were at the lake and terrestrial animals were more important at 
lakeside sites that previously thought. Second, cattail was very heavily exploited 
when available, perhaps to the exclusion of other resources for that short time. 

At a gross level, the La Quinta coprolite data appear to be relatively homoge- 



sam 



exist in the clustering of constituents. Analyses of constituent distributions and 
clustering can add considerable detail to the understanding of human ecology and 

adaptation. 

By combining the analyses of multiple lines of dietary evidence, it is possible 
to record and analyze dietary patterns that provide considerable information 
regarding people in antiquity. In addition to general dietary constituents, the 
reconstruction of cuisine, pharmacology, and other patterns are possible, 
that this line of research has only begun. 



I hope 



ACKNOWLEDGEMENTS 



I appreciate the discussions and correspondence with Vaughn Bryant, Jr., Ken 



Sobolik 



M. Yohe II on this subject. W.I. Follett of the California Academy ot bciences proviaeu eun^u- 
erable information and guidance on the fish, Adella Schroth assisted greatly with the state- 



Gonzolo 



Th 



Annual Meeting of the Society for American Archaeology, Atlanta. 



LITERATURE CITED 



BEAN, JOHN LOWELL. 1978. Cahuilla. 
Pp. 575-587 in Handbook of North 
American Indians, Vol. 8: California. 
Robert F. Heizer (editor). Smithsonian 
Institution, Washington, D.C. 

CASTETTER, EDWARD F. and WILLIS H. 
BELL. 1951. Yuman Indian Agricul- 
ture. University of New Mexico Press, 
Albuquerque. 



CUMMINCS, LINDA SCOTT. 1989. 
Coprolites from Medieval Christian 
Nubia: An Interpretation of Diet and 
Nutritional Stress. Unpublished Ph.D. 
Dissertation, Department of An- 
thropology, University of Colorado, 

Boulder. 
FARRELL, NANCY. 1988. Analysis of hu- 
man coprolites from CA-RIV-1179 and 



14 



SUTTON 



Vol. 13, No. 1 



CA-RIV-2827. Pp. 129-142 in Archae- 
ological Investigations at CA-RIV-1179, 
CA-RIV-2823, and CA-RIV-2827, La 
Quinta, Riverside County, California. 
Mark Q. Sutton and Philip J. Wilke 

(editors). Coyote Press Archives of 
California Prehistory No. 20, Salinas, 
California. 

LLETT, WILLIAM I. 1988. Analysis of 
fish remains from archaeological sites 
CA-RIV-1179 and CA-RIV-2327, La 
Quinta, Riverside County, California. 
Pp. 143-155 in Archaeological Investi- 
gations at CA-RIV-1179, CA-RIV-2823, 
and CA-RIV-2827, La Quinta, Riverside 
County, California. Mark Q. Sutton 
and Philip J. Wilke (editors). Coyote 
Press Archives of California Prehistory 
No. 20, Salinas, California. 



FRY. 



The 



Prehistoric Diets. Robert I. Gilbert, Jr. 
and James H. Mielke (editors). Aca- 
demic Press, Orlando. 



GOODMAN 



RIV 



Goodman and Brooke 



Arkush: Archaeological Investigations 



RIV 



in Central Riverside County, Cali- 
fornia. Report on file at the Eastern 
Archaeological Information Center, 
University of California, Riverside. 
__ and BROOKE S. ARKUSH. 



1990. Archaeological Investigations at 
CA-RIV-3793, Located North of Indio 
in Central Riverside County, Califor- 
nia. Report on file at the Eastern Ar- 
chaeological Information Center, Uni- 
versity of California, Riverside. 

HYLAND, DC, J.M. TERSAK, J.M. 
ADOVASIO, and M.I. SIEGEL. 1990. 
Identification of the species of origin of 
residual blood on lithic material. 
American Antiquity 55:104-112. 

JONES, A.K.G. 1986. Fish bone survival 
in the digestive systems of the pig, 
dog, and man: Some experiments. 
Pp. 53- 61 in Fish and Archaeology: 
Studies in Osteometry, Taphonomy 
Seasonality, and Fishing Methods. 
D.C. Brinkhuizen and A.T. Clason 
(editors). British Archaeological Re- 
ports International Series 294, 
Oxford . 

MCGINNIS, SAMUEL M. 1984. Fresh- 



water Fishes of California. University 
of California Press, Berkeley. 
MUNZ, PHILIP A. 1974. A Flora of South- 
ern California. University of California 

Press, Berkeley. 

NEWMAN, MARGARET E., ROBERT M. 
YOHE II, HOWARD CER1, and MARK 
Q. SUTTON. 1993. Immunological pro- 
tein analysis of non-lithic archaeological 
materials . Journal of Archaeological Sci- 
ence 20:93-100. 

REINHARD, KARL J. and VAUGHN M. 
BRYANT, JR. 1992. Coprolite analysis: A 
biological perspective on archaeology. 
Pp. 245-288 in Archaeological Method 
and Theory 4. Michael B. Schiffer 
(editor). University of Arizona Press, 
Tucson. 

SHAFER, HARRY J., MARIANNE MAREK, 
and KARL J. REINHARD. 1989. A Mim- 
bres burial with associated colon re- 
mains from the NAN Ranch ruin, New 
Mexico. Journal of Field Archaeology 

16 : 17-30 . 

SHREVE, FORREST and IRA L. WIG- 
GINS. 1964. Vegetation and Flora of the 
Sonoran Desert, Two Volumes. Stan- 
ford University Press, Stanford. 

SOBOLIK, KRISTIN D 1988a. The impor- 
tance of pollen concentration values 
from coprolites: An analysis of south- 
west Texas samples. Palynology 12:201- 
214. 

1988b. Diet change in the 

Lower Pecos: Analysis of Baker Cave 
coprolites. Bulletin of the Texas Archae- 
ological Society 59:111-127. 

, 1990. A nutritional analysis 



of diet as revealed in prehistoric 
human coprolites. The Texas Journal of 

Science 42i23 — 36. 
SUTTON, MARK Q. 1988. CA-RW-1179: 
Site description, research focus, field 
method, stratigraphy, features, and 
dating. Pp. 37-52 in Archaeological 
Investigations at CA-RIV-1179, CA-RW- 



Q 



Q 



Wilke 



Coyote Press Archives of California 
Prehistory No. 20, Salinas, California. 

and ROBERT M. YOHE II- 

1988 Terrestrial and avian faunal re- 



RIV 



Archaeological Investigations at CA-RIV- 
1179, CA-RIV-2823, and CA-RIV-2827, 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



15 



La Quinta, Riverside County, Califor- 
nia. Mark Q. Sutton and Philip J. Wilke 
(editors). Coyote Press Archives of Cal- 
ifornia Prehistory No. 20, Salinas, Cali- 
fornia. 

SUTTON, MARK Q. and PHILIP J. WILKE 
(editors). 1988a. Archaeological Investi- 
gations at CA-RIV-1179, CA-RIV-2823, 
and CA-RIV-2827, La Quinta, Riverside 
County, California. Coyote Press 
Archives of California Prehistory No. 
20, Salinas, California. 

1988b. Archaeological inves- 
tigations at the CA-RIV-2823 rock cairn 
complex. Pp. 15-20 in Archaeological 

Investigations at CA-RIV-1179, CA- 
RIV- 
2823, and CA-RIV-2827, La Quinta, 
Riverside County, California. Mark Q. 
Sutton and Philip J. Wilke (editors). 
Coyote Press Archives of California 
Prehistory No. 20, Salinas, California. 

SWOPE, KAREN K. 1988. Plant remains 
recovered by flotation from CA-RIV-1179. 
Pp. 119-128 in Archaeological Investiga- 
tions at CA-RIV-1179, CA-RIV-2823, and 
CA-RIV-2827, La Quinta, Riverside 
County, California. Mark Q. Sutton 
and Philip J. Wilke (editors). Coyote 
Press Archives of California Prehistory 
No. 20, Salinas, California. 

WALLACE, WILLIAM J. 1978. The chuck- 
walla: A Death Valley Indian food. 
The Journal of California Anthropology 
5:109-113. 



WATSON, PATTY JO. 1974. Theoretical and 
methodological difficulties in dealing 
with paleofecal material. Pp. 239-241 
in Archaeology of the Mammoth Cave 
Area. Patty Jo Watson (editor). Aca- 
demic Press, New York. 

WEIDE, MARGARET L. 1976. A cultural 
sequence for the Yuha Desert. Pp. 81- 
94 in Background to Prehistory of the 
Yuha Ddesert Region. Philip J. Wilke 
(editor). Anthropological Papers, No. 
5, Ballena Press. 

WILKE, PHILIP J. 1978. Late Prehistoric 
Human Ecology at Lake Cahuilla, 
Coachella Valley, California. Contribu- 
tions of the University of California 
Archaeological Research Facility No. 

38, Berkeley. 

___^ and MARK Q. SUTTON. 1988. 
Summary and inferences. Pp. 157-164 
in Archaeological Investigations at CA- 
RIV-1179, CA-RIV-2823, and CA-RIV- 
2827, La Quinta, Riverside County, Cal- 
ifornia. Mark Q. Sutton and Philip ]. 
Wilke (editors). Coyote Press Archives 
of California Prehistory No. 20, Sali- 
nas, California. 

YOHE, ROBERT M. II. 1990. Archaeological 
investigations at five sites located at 
one eleven La Quinta Center in the city 
of La Quinta, central Riverside 
County, California. Report on file, 
Eastern Archaeological Information 
Center, University of California, 
Riverside. 



BOOK REVIEW 



Delf ina Cuero: Her Autobiography: An Account of Her Last Years and Her Ethno- 



botanic Contributions. Florence Connolly Shipek. 



Menlo 



$19.50 (hardcover), $12.50 (paperback). lbbN 
0-87919-123-6 (hardcover), 0-87919-122-8 (paperback). 

As Ballena Press Anthropological Paper No. 38, this small book includes 
Delfina Cuero's autobiography, which was published originally in 1968, plus such 

nm.r i. • i . r i w_ r m^Q 4-^ 1Q79 a n^rHal lktinff of her 



material 



ii^w material as an account or ner lire rrom ivoo iu ±^/^, a K «*n«. »~*~» — --— 
ethnobotanical contributions, two photographs, and a map of southwestern Cal- 
ifornia and adjacent Baja California. Delfina Cuero was a Diegueno Indian, or 
Kumeyaay, who was born about 1900 and lived in San Diego County and nearby 
northern Baja California Norte until her death in 1972. Her perspective on the use of 
and change in the landscape and natural resources of the area as well as the data on 

specific plants are the two distinct contributions of major interest to ethnobotanists. 
-T-.1 . „ .« . . .• ~~*a rf^^rro nf r>l^nt<; and uni- 



mportance 



16 



BOOK REVIEW Vol. 13, No. 1 



mals to the survival 



tion period is clearly stated. The effect of cultural ignorance and violation of legal 



reements 



exam 



gathering territories by land developers and two countries, and the negative 
impact of cessation of female initiation ceremonies on such activities as traditional 
plant use and human reproduction. The correlation is evident between cultural 
disintegration and habitat loss. 



The 



Kumeyaay 



Marsh 



from Delfina Cuero during separate visits to Miss 



both in coastal southern California. It is included because she was the only person 



testimony to her contribution. Most 



the entries have basic information but four plants have no use listed and 15 are 



without native names. 
The 



with multiple uses. The ethnocentric categories are broken down as follows: medi- 



adornment 



smoking substance 1, thirst reliever 1, animal food 1, fiber 1, and construction 
material 1. Unfortunately, no voucher specimens were obtained. Plant identifica- 
tion was made in the field using P. A. Munz' A Flora of Southern California (1974) 
although this book was published after the field trips. In a few cases, the tax- 
onomic nomenclature of the ethnobotanical list does not correspond to that in the 



mentation 



may 



name 



// 



fatua 



unidentified semi-domesticated grain). The lack of voucher specimens and the 



them many years later may 



woman 



are interested in the reliability of this information. None-the-less Florence Con- 



Mrs 



Mrs. Rosalie Pinto Robertson. Dr. Mar 



Kumeyaay collaborator. Most 



comp 



special appreciation of Delfina Cuero for having shared her life and ethno- 
botanical experiences with us. The historical perspective in this book puts into 
painful perspective the destruction of a people, their knowledge and their plants. 



LITERATURE CITED 



MUNZ 



ifornia. University of California Press, 
Berkeley. 



Robert Bye 



Jard 



Autonoma de Mexico 



Mexico, DF, MEXICO 



/. Ethnobiol. 13(1): 17-54 



Summer 1993 



NEW PERSPECTIVES ON A WILD GOURD 

IN EASTERN NORTH AMERICA 



C. WESLEY COWAN 

Curator of Archaeology 

Cincinnati Museum of Natural History 

1720 Gilbert Ave. 
Cincinnati, Ohio 45202 



BRUCE D. SMITH 

Curator of North American Archaeology 

National Museum of Natural History 

Smithsonian Institution 
Washington, D.C. 20560 

ABSTRACT.— Free-living Cucurbita pepo gourds have been documented in differ- 
ent areas in the temperate eastern woodlands since the first few decades of the 
19th century. Until recently, however, little attention has been afforded these 
populations, and their potential role in the evolution of domesticated forms of pepo 
squashes has been obscured. This paper focuses on the botanical history of these 
free-living gourds, examines the availability of cultivar ornamental gourds to 19th 
century American gardeners, and summarizes habitat information for contempo- 
rary populations in several areas in the East. Important morphological measure- 
ments from both contemporary free-living and cultivar pepo gourd populations as 
well as archaeological specimens are compared. Combined, these data suggest 
that free-living Cucurbita pepo gourds are ancient members of an eastern flora, and 
could well be the progenitors of some domesticated squashes. 

RESUMEN.- Desde las primeras decadas del siglo XIX se ha documentado la 
existencia de calabazos no cultivados de la especie Cucurbita pepo en diferentes 
areas de los bosques templados del este de los Estados Unidos de Norteamerica. 
Sin embargo, hasta hace poco se les habia otorgado poca atencion a estas pobla- 
ciones, y su papel potencial en la evolucion de las formas domesticadas de la 
calabaza pepo ha sido dificil de discernir. Este trabajo se enfoca a la histona 
botanica de estos calabazos no cultivados; examina la disponibilidad de cultivares 
de calabazos ornamentales para los jardineros norteamericanos en el siglo XIX, y 
resume la informacion sobre el habitat de las poblaciones contemporaneas en 
varias areas del oriente de los Estados Unidos. Se comparan mediciones mor- 
fologicas importantes tanto de poblaciones contemporaneas cultivadas y no culti- 
vadas de calabazos pepo, como de especimenes arqueologicos. Estos datos com- 
binados sugieren que los calabazos no cultivados de Cucurbita pepo son miembros 
antigous de la flora del este de Norteamerica, y bien podrian ser los progemtores 
de algunascalabazas domesticadas. 

RESUME.- L'existence des gourdes sauvages Cucurbita pepo a ete documentee 
dans plusieurs endroits des regions boisees temperees de l'Est depuis les pre- 



18 



COWAN 



Vol. 13, No. 1 



mieres decenies du 19eme siecle. Jusqu'a recemment toutefois, on a prete peu 
d'attention a ces populations locales, et leur role possible dans revolution des 
formes cultivees de la courge pepo n'a pas ete suggere. Cet essais est consacre a 
l'histoire botanique de ces gourdes sauvages. II examine l'acces des jardiniers 
americains du 19eme siecle a la forme cultivee des gourdes ornementales, et 
synthetise les informations sur l'habitat des populations contemporaines dans 
plusieurs regions de l'Est. D'importantes mesures morphologiques des especes 
sauvages et des populations cultivees de la gourde pepo contemporaine ainsi que 
des specimens archeologiques sont compares. Vues dans leur ensemble, ces donnees 
suggerent que les gourdes Cucurbita pepo sauvages sont des ancients membres de 
la flore orientale, et pourraient parfaitement avoir ete les progeniteurs de cer- 
taines especes de courges cultivees. 



. . . Is this Texas cucurbit a garden escape as suggested by Gray, or is it the proto- 
type from whence came the numerous cultivated forms of Cucurbita pepo? Is it an 



indigene or a foreigner? 



(A.T. Erwin 1938:253) 



INTRODUCTION 



In a little recognized and rarely referenced article, A.T. Erwin first posed a 



importance in documenting < 
I plant domestication. As mi 



interven 



ing years, but it still involves the status of the Texas wild gourd and related free 



America 



and 



domestication 



pepo in the eastern woodlands. 1 



streams 
survive 



tion) are being documented in increasing numbers in other areas of eastern North 
America. In addition, since the 1980s, rind fragments and seeds of a similar gourd 
have been recovered from archaeological deposits in eastern North America that 
predate the earliest evidence of domestication of indigenous seed plants by 2500 
years (Smith et al. 1992; Cowan 1990). Over the past decade both these present- 
day and prehistoric gourds have been the topic of considerable discussion. Do the 
early Cucurbita pepo gourds of the East that predate the circa 4500-3500 B. P. domes- 
tication of local plants represent a wild plant indigenous to the region, or an early 
domesticate introduced from Mexico? If these pre-4500 B.R gourds can be traced 
back to Mesoamerica, then agricultural developments in the East might be consid- 
ered derivative rather than independent, since a tropical crop plant would have 
arrived in advance of domestication of local species. On the other hand, if these 
early cucurbits represent an indigenous wild gourd, then there can be little doubt 
that eastern North America could have been an independent center of pla nt 
domestication. The debate regarding these early gourds has added new signifi- 
cance to a long-standing discussion of present-day free-living C. pepo gourd popu- 
lations in eastern North America: do thev represent an indigenous wild gourd, or 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 19 



the recent escape from cultivation of a domesticate that ultimately originated in 
Mesoamerica? 

Intertwined with and overlaying these questions is the possibility that C. pepo 
was independently domesticated in eastern North America a full 5000 years after 
it was first domesticated in Mexico (Flannery 1986). Based on available archae- 
ological evidence for the earliest clearly domesticated C. pepo in the East (4000- 
3000 B.P.), this second episode of domestication would have occurred at the same 
time as other seed crops were brought under domestication. 

Recent biochemical analyses of cultivar and wild populations of Cucurbita pepo 
lend support to the role a wild eastern squash may have had in the domestication 
process. Isozyme studies indicate two distinct lineages characterize C. pepo (Decker- 
Walters 1990; Decker 1988; Decker and Wilson 1986, 1987; Wilson 1990). One of 
these seems to have evolved in Mexico, the other in eastern North America. The 
Mexican lineage contains the marrows and pumpkins while the eastern counter- 
part includes the crooknecks, acorns, and most cultivated ornamental gourds. 
The free-living gourds from various areas in the southeast are also part of this 
lineage. And like many other domesticated members of the lineage, the free-living 
gourds are vining, monoecious annuals that produce large insect pollinated 
flowers. They are easily recognizable in the wild by their distinctive sinuously lobed 
leaves, yellow flowers, and ivory or green striped fruits. 

Elsewhere (Smith et al. 1992) we have summarized the theoretical arguments 
of the proponents of a single Mesoamerican origin for C. pepo and those who 
believe C. pepo squash may have been domesticated separately in Mexico and 
eastern North America. Our purposes in this paper are fourfold: (1) to document 
the geographical distribution of modern free-living gourd populations in the East; 
(2) to establish the historic time depth of these gourds; (3) to describe their niche 
and habitat preferences in the Ozarks of Missouri and Arkansas; and (4) to describe 
the morphological characteristics of fruits of these populations. The following 
discussion will make it clear that wild gourds are not limited in distribution to 
Texas, that the more widely distributed gourd represents an ancient member of an 
eastern flora, and that the ancestral relationship of this wild gourd to cultivated 
forms of Cucurbita pepo can be partially untangled by a comparison of modern and 
archaeological collections (see also Decker- Walters et al. and Newsom et al. this 
volume). 



DOCUMENTING 



MIDWEST 



Free-living populations of Cucurbita gourds in eastern North America have 
been collected for over 150 years. In addition to those in eastern Texas recognized 
in the 1830s, specimens were also collected in the 1840s as far north as St. Louis, 
Missouri AHK^,,r,K fV, Q c^iA^ cf„t„ c nf tho nlant fa uncertain, wild gourds may 



peninsul 



quarter of the nineteenth century (see below). By and large, however, with the 



i 



profes 



20 



COWAN & SMITH Vol. 13, No. 1 



forms 



assumed 



Until the mid-1980s relatively little attention was focused on C. pepo gourd 
populations outside of Texas, likely because they were assumed to be ephemeral 
escapes from cultivation (Asch and Asch 1985:157; Heiser 1985:16-17). Julian 
Steyermark appears to have played an important role in establishing that free- 



itimate 



from southwest Missouri 



map 



>/ Missouri (Steyermark 



seems 



exam 



absent from a 1955 survey (Jones and Fuller 1955) yet is documented as present in 
seven counties in the next published atlas (Mohlenbrock and Ladd 1977). Simi- 
larly while it goes unmentioned in earlier surveys, "C. pepo ssp. ovifera" is 
recorded in 10 Arkansas counties in Edwin Smith's 1978 state atlas (E. Smith 1978, 
1988), with many of the relevant herbarium specimens collected in the mid-1970s. 

Another important impetus to the collecting and documentation of free-living 
gourd populations outside of Texas came in the mid-1980s when Decker- Walters 
included gourds from Illinois, Arkansas, and Alabama in her dissertation research 
on the taxonomy and evolution of C. pepo and proposed that these var. texana-like 
geographical outliers might represent relict populations of a wild indigenous 
gourd (Decker 1986). As a result of Decker-Walter's inclusion of these specimens, 
Heiser added them to his distribution map for var. texana, expanding it to encom- 
pass the "distribution of C. texana and plants approaching C. texana" (Heiser 
1989:473). Similarly Michael Nee included eastern free-living outliers recorded 
for Missouri (Steyermark 1963), Illinois (Mohlenbrock and Ladd 1977), Arkansas 
(E. Smith 1988), and Alabama on his geographical range 
the caption note that "some dots outside of Texas may rep 

1990:61). Decker-Walter's research, and the subsequent mapping efforts of Heiser 
and Nee, marked an important shift in that they considered both Texas and "out- 
lier" populations of var. texana and began to chart the full geographical distribu- 
tion of free-living Cucurbita pepo gourds in the southern and eastern United States. 

Building on these studies, we began a more concerted effort in the fall of 1990 
to establish the range of free-living Cucurbita pepo gourds. Herbaria were canvased 
regarding accessions of "escapes" of Cucurbita (often listed as C. pepo ssp. ovifera), 
colleagues were questioned regarding sightings of free-living gourds, and an initial 



map 



survey ot drainages in the eastern Missou 



ementing the surveys carried out by Michael 



Smith 



1992). The initial results of this distributional study are shown in Fig. 1 and sum- 
marized in Table 1. Outside of Texas, 26 county records were added to the 29 
previously documented, and fruits were collected from over 30 locations in seven 
states (Table 1). Within Texas, gourd populations were documented in 20 counties. 
Based on our survey the primary present-day geographical range of free- 
living Cucurbita pepo gourds west of the Mississippi River extends in a broad north- 
south band from south-central Texas to central Illinois. Stretchine for more than 



Summer 1993 



JOURNAL OF ETHNOBJOLOGY 



21 




FIG. l.-Current distribution of free-ranging Cucurbita pepo gourds in eastern 



America 



denote counties where field collections were made in 1990-91. Lighter shaded 
areas denote counties where herbaria records exist. 



1,400 kilometers (900 miles) this north-south band appears to be divided into five 
subareas. In the large Texas subarea populations are documented througnou 
much of the drainage catchment of the eastern part of the state, occurring in ail ot 
the southeastern flowing rivers that enter the gulf of Mexico between Corpus 
Christi and Galveston (the Nueces, San Antonio, Guadalupe, Colorado Brazos, 
and Trinity rivers). An absence of populations to date from the Neches and Sabine 
drainages separates the Texas subarea from populations documented along the 
Red River and lower Ouachita River in Arkansas. A similar absence of recorded 
populations within the Ouachita mountains of west-central Arkansas separates 



22 



COWAN & SMITH 



Vol. 13, No. 1 



TABLE 1.— Geographical distribution, by county, of free-living Cucurbita pepo 
gourds in eastern North America. 1 



County 



ALABAMA 

Dallas 

Greene 

Greene 

Marengo 

Mobile 

Monroe 

Wilcox 

Wilcox 

ARKANSAS 

Ashley 

Benton 

Benton 

Benton 

Faulkner 

Hempstead 

Independence 

Independence 

Izard 

Miller 

Newton 

Prairie 

Searcy 

Searcy 

Stone 

Woodruff 

ILLINOIS 

Cass 

Coles 

Douglas 

Douglas 

Jackson 

Jersey 

Madison 

Massac 

Morgan 

Piatt 

Randolph 
St. Clair? 
St. Clair 
St. Clair 
St. Clair 
St. Clair 
St. Clair 
St. Clair 
Union 



Collector 



Year Herbarium 



C. Sheldon 
M.L. Roberts 



1988 

1982 Alabama 



R. Deramus 
C. Sheldon 
R. Haynes 
C. Sheldon 



1966 Alabama 

1988 

1978 Alabama 

1988 



M. Hoffman 

D. Dickson 

E. McCollum 
D. Oliver 

S. Harrison 



1990 
1990 
1990 

1975 Arkansas 

1976 Arkansas 



Smith and Cowan 



1990 



D. Oliver 

B. Hinterthuer 

Smith and Cowan 



1975 Arkansas 

1977 Arkansas 
1990 



Smith and Cowan 



1990 



G. Jones 
G.Jones 



1966 Illinois 
1966 Florida 



G . Fritz 



1990 



R. Mohlenbrock 



1986 



W. Welsch 
H. Eggert 
H. Eggert 
H. Eggert 
H. Eggert 
H. Eggert 
J. Kellogg 



1862 Illinois 

1875 Missouri 

1876 Texas 

1892 Missouri 

1893 Harvard 
1893 Missouri 
1901 Missouri 



Reference 



this study 



Decker 1986 
Decker 1986 



this study 



this study 



Smith 1988 
this study 
this study 
this study 
Smith 1988 
Smith 1988 
Smith 1988 
this study 
Smith 1988 
Smith 1988 
Decker 1986 
Smith 1988 
Smith 1988 
this study 
Smith 1988 
this study 



Mohlenbrock and Ladd 1977 
Mohlenbrock and Ladd 1977 



Mohlenbrock and Ladd 1977 

this study 

Mohlenbrock and Ladd 1977 



Mohlenbrock and Ladd 1977 
Mohlenbrock and Ladd 1977 
Decker 1986 



Mohlenbrock and Ladd 1977 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



23 



County 



Collector 



Year Herbarium 



Reference 



KENTUCKY 

Powell 

LOUISIANA 

Bossier 
St. Helena 
Tensas 
W. Feliciana 

MISSISSIPPI 

Claiborne 

Forrest 

Rankin 

MISSOURI 

Barry 

Christian 

Douglas 

Greene 

Howell 

McDonald 

Newton 

Ozark 

Polk 

St. Louis 
St. Louis 
St. Louis 
St. Louis 
St. Louis 
Taney 
Texas 
Wright 

OKLAHOMA 

Adair 

Cherokee 
Mayes 

TEXAS 

Bell 

Brazos 

Brazos 

Burleson 

Calhoun 
Comal 

DeWitt 

DeWitt 

DeWitt 

Fayette 
Goliad 

Gonzales 
Gonzales 
Grimes 



W. Booth 



1990 



this study 



L. Baker 
C. Allen 

G. Fritz 

A. Martin 



1990 

1971 LSU 
1990 

1972 LSU 



this study 



this study 



K. Rogers 
K. Rogers 
S. Jones 



1978 Tennessee 
1971 Tennessee 
1970 Georgia 



J. Steyermark 



J. Steyermark 
Smith and Cowan 



1955 Missouri 

1990 

1957 Georgia 

1990 



Smith and Cowan 
Smith and Cowan 
G. Engelmann 
G. Engelmann 
Muehlenbach 
Muehlenbach 
Muehlenbach 



1990 
1990 



J. Steyermark 
Smith and Cowan 



M. Hoffmann 
B. Meyer 
D. Dickson 



1990 
1990 
1990 



Mahler 1988 
D.S. Correll 



Harman and Smith 
Lindheimer 
D.S. Correll 
D.S. Correll 
Tharp 



Steyermark 1963 
this study 
Steyermark 1963 
this study 
Steyermark 1963 
Steyermark 1963 
Steyermark 1963 
this study 
this study 



1846 Missouri Steyermark 1963 

1846 Missouri 

1961 Missouri 

1964 Missouri 

1972 Missouri 

1990 

1956 Harvard 

1990 



this study 
Steyermark 1963 

this study 



this study 
this study 
this study 



Texas 



Decker 1986 
Decker 1986 



Texas 
Texas 
1962 SWLS 

Texas 
Texas 



Decker 1986 
Decker 1986 
Decker 1986 

Decken 1986 
Decker 1986 



24 



COWAN & SMITH 



Vol. 13, No. 1 



TABLE 1.— Geographical distribution, by county, of free-living Cucurhita pepo 
gourds in eastern North America. 1 (continued) 



County 



Collector 



Year Herbarium 



Reference 



TEXAS (continued) 



Lee 

Madison 

Refugio 

Robertson 

San Jacinto 

San Patricio 

Sutton 

Travis 

Travis 

Travis 

Travis 

Washington 



Decker 1986 
Decker 1986 
Decker 1986 
Decker 1986 
Decker 1986 
Jones 1975 



Reed 

A.T. Erwin 
Barkley 
Tharp 

Strandtmann 



Texas 



1938 



Erwin 1938 



Texas 

Texas 
Texas 



Decker 1986 



Requests for information regarding free-living C. pepo specimens were sent to the following herbaria: 
University of Alabama, Auburn University, University of Arkansas, University of Florida, Florida 
State University, University of Georgia, Emory University, University of Illinois, Southern Illinois 
University University of Kentucky Indiana University, DePauw University, Louisiana State Univer- 
sity, Southwest Louisiana State University, Harvard University, University of Mississippi, University 
of Missouri, Missouri Botanical Gardens, University of North Carolina, Duke University, University 
of Cincinnati, The Charleston Museum, The University of Tennessee, The University of Texas, Vir- 
ginia Tech University, The University of West Virginia, National Museum of Natural History, Smithso- 
nian Institution. 



surv 



Ozark Plateau of southern Missouri and northern Arkansas. Within the Ozark 
Plateau, populations have been documented in small western trending streams 
and larger southern flowing rivers of the drainage system of the Arkansas River, 
as well as in north-flowing streams of the Osage and Gasconade systems, and 
along the east-southeastern flowing Buffalo-White River drainage system. About 
150 kilometers separates the Ozarks subarea from the northernmost, Illinois sub- 

the lower Illinois River, the upper reaches of the Kaskas- 



encompasses 



mile) portion of the Mississippi 



section of the lower Ohio River. A fifth subarea can be recognized along the central 
coastal plain of the Gulf of Mexico, with populations recorded along a number of 
major and minor drainage systems emptying into the Gulf, including those of the 
lower Mississippi, Pascagoula, Pearl, Mobile-Tombigbee, and Alabama Rivers. 

While it is not nossiblp tn n^rprf^in fViP H^crr^o fr» \A>Vii^fi fhp»cp annarent inter- 



survey and collection, planned field research in these intervening areas, along 
with ongoing comparative genetic analysis of the populations of different sub- 
areas (see Decker-Walters et al, this volume), will provide a clearer picture of the 
degree of extant geographic and genetic separation that does exist. It would be 



nonuniform 



J r f ww ^ w ^**/ "*~ v v-ai t_lVV-ri 11 l\_ 1 IKJl IC4I IIX Wl III U1JU IL'UUVi • — - 

mented populations, that this eastern free-living Cucurbita pepo gourd today has a 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 25 



largely unbroken distribution in river systems from south Texas as far north as 
central Illinois, and along the Gulf coastal plain from Corpus Christi to Mobile. 



CURRENT EXPLANATIONS FOR THE DISTRIBUTION 

OF FREE-LIVING Cucurbita pepo GOURDS 



time 



determine with certainty. Although gourds were observed in 1835 in Texas, their 
presence was not noted in the Ozarks until the mid-1950s, they seemingly went 
unrecorded in Illinois and along the Gulf coastal Plain until the mid-1960s, and 
were only documented along the Red River in the mid-1970s. 

The distribution of gourd populations and the sequence of their initial colleo 



some 



World War 



mo 



as occupying an extremely narrow niche— that of cultivated floodplain fields— in 
highly specialized agroecosystems. 

There is little question that Cucurbita pepo gourds have become a problem weed 



The 



become 



them 



herbicide effectiveness reported density counts ranging from 32 plants per square 
meter at Fulton, Arkansas, to 43 per square meter at Conway, Arkansas, to 129 



meter 



, M _ ___-living pepo gourds were observed in an "infested" bot- 
tomland cotton and soybean field located between the levee and Mississippi River 
near St. Joseph, in Tensas Parish, Louisiana (Gayle Fritz, personal communica- 
tion, 1991). On a much smaller scale, several dozen gourds were collected in 
November of 1990 from fallow floodplain fields along Willow Chute, an old chan- 
nel of the Red River just north of Shreveport (Bossier Parish) in northwest Loui- 
siana (Frank Schambach, personal communication, 1990). In Kentucky, a free- 
living Cucurbita pepo gourd known locally as "Johnny Gourd" has grown in the 
Ohio and Green river floodplain fields of corn and soybeans of Union and Hen- 



for years. When questioned abo 
Western Kentuckv most farmers 



Most 



World War 



In our view a likely explanation for the seemingly shallow time depth for free- 
ranging Cucurbita pepo gourds outside of Texas may be explained by: (1) a lack of 
interest by botanists, and (2) the plant may be aggressively re-establishing itseli in 



sissippi River. 



The Asches (1992) have also suggested that Texas wild gourd may have ex- 
panded its range to the east and north over the past 40 years as the result o 



commerce 



gourd seeds. These potential human vectors of dispersal, along with other pos- 
sibilities, could well be involved in the continuing expansion of the range ot tree- 



26 



COWAN & SMITH Vol. 13, No. 1 



living pepo gourds in eastern North America. The suggestion that seeds of the wild 



Texas gourd were packaged and sold as ornamental gourds does, in fact, provide a 
context for the subsequent "escape" of an "ornamental gourd" across the East, but 
such events, if they occurred, would clearly have to be considered human medi- 
ated range extension of a still wild plant, rather than the escape of a domesticated 
plant. A related and intriguing question involves the possible existence of long 
established relict populations of Cucurbita pepo gourds outside Texas in areas mini- 
mally disturbed by humans and agroeconomies . Does the ongoing range exten- 
sion of a wild gourd in the East have its origins only in Texas populations, or are 
there other source areas where relict populations have existed, largely unnoticed, 



time? 



m 



Mississippi Valley, 



miss 



ample reason to understand why botanists would have ignored the plants. As long 
as they were considered an "escape" from cultivation, there was little reason to 



from 



completely indigenous cucurbit remained 



world for hundreds of years . The Okeechobee gourd (Cucurbita okeechobeensis) was 



While Bartram described a seem 



(Small 



(Harper 1958), it is impossible to determine whether this was the Okeechobee 
or the Texas gourd (Walters and Decker-Walters 1993). The fact that this plant 
escaped the attention of professional botanists is probably also a reflection of the 



sim 



There are three straightforward tests of whether this ongoing range expansion 



establishment 



gourd in previously occupied territory, or the spread of a new weed into a narrow 



agroeconomy. 



(1992) new narrow niche floodplain field weed explanation. Especially significant 
in this regard are any indications that gourds were present beyond Texas prior to 
the post- World War II appearance of the soybean field agroecosystem habitat into 
which the ornamental gourds hypothetically escaped. Presence of free-living gourds 
in eastern North America prior to the late 1940s would clearly undermine the new 
soybean field habitat hypothesis. 



ntemporary 



domesticated forms of orname 



Midwest are garden escapes that have become naturalized. While suggesting this 
occurred after World War II, they do not specify which ornamentals escaped. It is 
important, then, to understand (1) when ornamental gourds began to be widely 
available to home gardeners, (2) what types were marketed, and (3) the context of 
their cultivation. If it can be shown, for example, that numerous ornamental gourd 



World War 



extremely limited 



of free-living gourds might be regarded as evidence of a truly wild Cucurbita pepo 



mechanized 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 27 



A third test focuses on how tightly modern gourds are restricted to agri- 
cultural settings. Are they narrowly adapted weeds restricted to soybean field 
settings as proposed by the Asches? Are they tightly tied to a companion wood 
role for domesticated C. pepo? Or are they more generally adapted weeds of the 
agroecosystem? Do these gourds occur only in human-disturbed habitats, or do 
they also exist in natural settings, either close to or far removed from gardens and 
agricultural fields? Obviously the further removed from agricultural contexts that 
such gourds grow, the stronger the case for the range extension of a wild gourd. 



more 



roecosystem 



DOCUMENTING THE HISTORIC TIME 

THE WILD Cucurbita GOURD TM TH 



While 



World War 



America, a growing body of archaeobotanical data suggests a wild morphotype 
Cucurbita pepo gourd was present in numerous places in the Southeast and Mid- 
west as early as 12,000 years ago, to as late as 200 years ago. 

We have already noted the ancient presence of a morphologically wild cucur- 
bit in eastern North America. This evidence places a thin-walled, brittle-shelled, 
small seeded Cucurbita pepo gourd in Florida by 12,000 B.P. (Newsom et al., this 
volume), in west-central Illinois by 7000 years ago (Conard et al. 1984), in eastern 
Tennessee 6990-5300 B.P. (Crites 1991), in south-central Missouri and eastern Ken- 
tucky by 4500 years ago (Kay et al. 1980, Cowan 1990), and as late as A.D. 800-1750 
in eastern peninsular Florida (Decker and Newsom 1988). As yet undated, but 
morphologically wild cucurbits have also been recovered from several Ozark rock- 
shelters (e.g., Gilmore 1931, Plate 24a). Unequivocal evidence of domesticated 
forms, indicated by an increase in rind thickness and seed length, does not seem 
to occur in the East until sometime after 4000 B.P. (Cowan 1990; Smith 1992). 

In addition to the archaeological presence of a wild gourd in Florida for at least 
12,000 years, and across the Midwest and Southeast for the past 7000, our herbaria 
survey produced records of a wild morphotype Cucurbita pepo in the St. Louis area 
between 1846-1901 (Table 1). These collections-housed in the herbaria of the Uni- 
versity of Texas, Harvard University, the Illinois State Museum, and the Missouri 
Botanical Garden— provide dramatic evidence of how botanists have struggled 
over the past 150 years with the taxonomic puzzle represented by the wild gourd. 

Collections from the St. Louis area are especially important. As the gateway to 
the west in the mid-nineteenth century, and the largest city between the Mis- 
sissippi and the Pacific, St. Louis was a center of science and culture. It was also 
the staging area for every important scientific exploration of the western territo- 
ries, and was in every sense of the word, on the edge of civilization. 

Perhaps because of its strategic location, St. Louis was also a center of scien- 



1859, was the focus of much 



The Missouri 



and contains significant 



collections that have a direct bearing on the history of Cucurbita evolution in the 
East. Three of the earliest collections of an eastern Cucurbita pepo gourd are curated 



28 



COWAN & SMITH Vol. 13, No. 1 



at the Garden. All were deposited by George Engelmann, a St. Louis physician, 
amateur botanist, and the Garden's long-time chief curator. Engelmann published 
dozens of articles dealing with botany and was considered an authority on several 
genera. Today, his botanical notes occupy 60 "large books," which are part of the 



Engelmann 



Mis 



American 



communication, August, 1990). 



specimens 



seeds collected in Texas. One (MO-3265655) was identified by Engelmann as "Cucur- 
bita ovifera Lin. var. pyriformis, " and was collected in September, 1846 from a French 
plant collector, Nicholas Riehl (the specimen label reads "Cult, from Texas seeds by 
N. Riehl, St. Louis"). A specimen collected in July, 1848 (MO-3265676) is simply iden- 
tified as "Cucurbita" and carries the inscription "From Texas seeds cult, in St. Louis." 
The third specimen (MO-3265373) is perhaps the most significant. Labelled only 
Cucurbita" with the note "Naturalized, St. Louis along fences and fields," this 



// 



September, 1846, the same year and month 



'f 



MO-3265655— the plant grown 



// 



'* 



n 



designate it as C. ovifera var. pyrift 



plant to species level, 



taxonomic placement 



If Englemann's "naturalized" specimen 



wild Cucurbita pepo gourd growing in St. Louis it would be reasonable to consider 
it a likely "escape." However, there also exist eight other collections of potentially 
wild Cucurbita gourds from St. Clair County. Illinois made between 1875 and 1893. 



made 



American Bottom 



until his death in 1904, at which time his personal herbarium of more than 23,000 
specimens was purchased by the Missouri Botanical Garden. During his active 
career Eggert "... collected assiduously all around St. Louis for a considerable 
distance, and his collection probably represented the flora of this district better 
than any other ever made" (Spaulding 1909:252). 

Although considered an expert on the flora of the St. Louis area, Eggert pub- 
lished only one brief 16-page pamphlet. His Catalogue of the Phaenogamous and 
Vascular Cryptogamous Plants of the Vicinity of St. Louis, Missouri, published in 1891, 
however, has a particularly interesting entry. Eggert lists nearly 1,100 plants as 
being indigenous to the area, and as late as 1909 his inventory was considered "by 
far the most complete list of our plants which has yet appeared" (Spaulding 1909:252). 
Among the plants he lists as native is Cucurbita ovifera var. pyriformis. 

Eggert 's labels from his "Herbarium Americanum" contain few specific locality 
data, but labels for three of the five specimens of Cucurbita gourds do list habitat 
information. These include: "lowland" (MO-768398), "waste places" (MO-3265660), 
and "prairies and waste places" (Harvard-Eggert s.n.). 

Eggert identified three specimens as "C. ovifera var. pyriformis" (MO-768398, 

TX-Eggert s.n., Harvard-Eggert s.n.), one as "C. ovalis" (MO-739055), and one as 

Cucumis" (MO-3265660). The remaining specimens were simply identified as 
Cucurbita." 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 2" 



specimens 



(MO 



1893 (Harvard-Eggert s.n.)— contain small (< 8 cm in length), pyriform, green- 



terms 



wild morphotype gourds collected during the present study (see below). 

In addition to these nineteenth century collections, at least one specimen was 



American Bottom just after the turn of the century (MO 



specimen 



// /"* ' C tt 



\fera," was collected by John Kellogg in 
woods adjacent to Fish Lake in October, 1901. Fish Lake is a large cut-off meander 
lake in the American Bottom. 

Annotations of these early collections made by subsequent scholars provide 
some indication of the broader taxonomic caution that has obscured the possibility 
of a wild Cucurbita pepo gourd being present in the East. Interestingly L.H. Bailey 
examined the Eggert specimen ("C. pepo var. piriformis" housed in the University 
of Texas Herbarium and proclaimed it to be C. texana. Taxonomic reassessments of 
the Missouri Botanical Garden specimens were recently examined and annotated 
by Thomas C. Andres of the L.H. Bailey Hortorium between 1989-1990. 

It is often difficult to distinguish between the different domesticated and wild 
forms of C. pepo based solely on herbarium specimens, where only a small portion 
of the plant might be present. As a consequence, contextual information supplied 
on the label is often considered, as is the identity of the collector. Plants collected 
in the railroad yards of St. Louis and identified as "escapes" by Hugh Cutler, a 
noted cucurbit researcher, were accepted by Andres, for example. 

Applying current nomenclature, Andres annotated Engelmann's specimens 
grown from Texas seed and originally labelled C. pepo var. pyrifbrmis, as Cucurbita 
texana. The specimen Engelmann described as "naturalized along fence rows and 
waste places" (emphasis added) and identified as Cucurbita, was assigned to C. 
pepo by Andres. This is particularly interesting when the specimens collected by 
Eggert from the same area a few years later are considered. Sometime after the 
Garden acquired Eggert's collection, the specimen he identified as "Cucumis" and 
listed as growing in "waste places in East Carondolet," was re-identified by an 
unknown researcher as a domesticated ornamental gourd ("C. pepo var. ovifera"). 
This same specimen was re-identified in 1989 by Andres as "C. pepo ssp. texana." 

In spite of the fact that the remaining nineteenth century Cucurbita specimens 
in the Missouri Botanical Garden all bear habitat descriptions indicating collection 
outside of contexts of cultivation (i.e., "waste places," "fence rows," "low ground," 
"prairies"), and one contains a mature fruit that falls within the wild morphotype, 
Andres was sufficiently hesitant regarding their taxonomic status to identify each 
only as C. pepo. 

We would argue that earlier recognition of an indigenous eastern North Ameri- 
can Cucurbita pepo gourd has been hindered by the absence of an appropriate 
taxonomic category and label. Engelmann and Eggert employed the taxonomic 
label "C pepo var. piriformis" based on the pear-shaped fruit to refer both to plants 
grown from Texas seeds and plants found growing outside of cultivation in St. 
Louis. Subsequent annotation of these specimens has either assigned them to the 
Texas wild gourd (C. pepo var. ovifera ssp. texana), or more cautiously C. pepo. We 
propose a different reading of this early herbarium evidence. 

By 1846 (11 years after its discovery in Texas), botanists in the St. Louis area 



30 



COWAN & SMITH Vol. 13, No. 1 



were collecting a seemingly wild morphotype gourd indistinguishable from 
C. pepo ssp. ovifera var. texana, and later Eggert included it as a member of the local 
flora. Similar collections were made throughout the remainder of the nineteenth 
and first few years of the twentieth centuries. Rather than simply representing 
"escapes" from local gardens we believe these early specimens were of an indige- 
nous, wild C. pepo gourd. 

Our herbaria survey, along with archaeologically recovered gourds, provide 
strong support for the long-term presence of an indigenous wild gourd in eastern 
North America long before modern field agriculture. These data provide the first 
link in our argument countering the position that the weedy infestations of gourd 
in f loodplain fields in the East represents a post-World War II escape into a narrow 
agricultural weed niche. An examination of the availability of cultivated forms of 
Cucurbit a pepo ssp. ovifera to eastern North American gardeners provides another. 



HISTORY OF THE AVAILABILITY OF ORNAMENTAL GOURDS 

TO GARDENERS IN EASTERN NORTH AMERICA 



more 



Agricultural and National Horticultural Libraries provides a solid baseline against 
which the availability of ornamental gourds (C. pepo ssp. ovifera) to the American 
gardener can be measured. Descriptions of ornamental eourds in these catalogues 



recomm 



time horizon for any supposed "escape" that might have occurred. 



While William 



able for distribution in the late eighteenth century, seed catalogues did not become 
widely available until after the Civil War. Thev nroliferated as the nineteenth 



M'Mahon's Catalogue of A 



exa m i 



•fern (M'Mahon 1804:20). C. ovifi 



listed as the "egg-shaped gourd." Significantly, no other fancy gourds are offered. 
™ r " ' " egg" gourd in seed catalogues until the mid- 



mention 



almanacs and catalogues advertised ornamen 

(M'Mahon 



V~^ j J *~* -*- v-^ -*-->^ 1 .A- ▼ X X ▼ X Vt X m \*J XI X. \^/ X. *r m \—J V-' *r i V* i— » V* <— *^^ x ^ 

ing arbors, trellises, walls, and fences. Two varieties are invariably offered in the 
earliest catalogues: "bi-colored," and "orange" (cf. William Prince and Sons 1834- 
35:19; Hovey and Co. 1845:6; B.K. Bliss 1860:19). 

By the end of the Civil War, the number of ornamental gourd varieties listed in 
catalogues increased dramatically. Both Vick's Illustrated Guide and Floral Catalogue 
for 1866, as well Hiram Sibley's Garden, Field and Flower Seed Catalogue for 1879, for 
example, list seven Cucurbita pepo gourds. James Vick's Fifth Annual Catalogue of 
Seeds (1866) is especially significant. In addition to offering a variety of fancy 
gourds, he offered a gourd described as "Egg-formed, like the fruit of White Egg 
plant, very beautiful and new" ( Vick 1866:43, emphasis added). A nearly identical 
description is found in M. O'Keefe and Sons catalogue for 1868 (pg. 29). By 1871, 
companies in Illinois, Ohio, and New York were offering the egg gourd, along 
with other varieties, almost always listing them as mainly useful for their climbing 
ability and for covering trellises and arbors. 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 31 



While the "egg" variety is listed in many catalogues in these decades— often 
with the description "like the fruit of a white egg plant"— in the 1880s several 
catalogues began providing a separate, lengthy description for this ornamental. 
Burpee's Farm Annual for 1883 is typical of this new status. Listed as the "White 
Egg-Formed Gourd" (1883:18) it is also called the "Japanese Nest-Egg Gourd." 
The fruits are illustrated and described as: 



. . . they exactly resemble, in color, size, and shape the eggs of hens, and do not 
crack, and are uninjured by the by cold or wet, they make the very best nest eggs 



Although hardly used today, before the advent of electronic incubation artificial 
"nest eggs" were often placed in a nest to encourage the hen to lay more eggs and 
remain on the nest until they hatched. Glass eggs were typically used by those 
who could afford them. The egg gourd seems to have been marketed as an inex- 
pensive alternative. 

Nest egg gourds continued to be advertised in late nineteenth and early twen- 
tieth century catalogues. As commercial poultry hatcheries developed, however, 
their status began to change. The D.M. Ferry & Company (later the Ferry-Morse 
Seed Company), provides a good example of the evolution of this transformation. 
During the first three decades of this century, Ferry lists and illustrates no less than 
six ssp. ovifera gourds, including the "Japanese nest egg." By 1933, however, these 
distinct varieties began to be subsumed under the single heading "mixed" gourds. 
While continuing to tout the general utility of all forms of gourds for covering 
trellises and arbors, in the 1937 catalogue (p. 57), the "small fruited mixed" listing 



suggests that these are of "An assortment 

decoration and other ornamental purposes. 

today and marks the first time this specifiec 

To summarize, our examination of see 



// 



</< 



While 



become 



Throughout the period 1800-1880, the egg gourd was only one of a number of 
gourds advertised for sale. Of these, the bi-color, and orange ball are the most 



d orname 
number oi 
ornamental 



to cover trellises and arbors. There is no indication they were marketed for any 
other use. This places the context of their cultivation in areas immediately adjacent 
to the home with no evidence that they were cultivated commercially. During the 
1880s, the so-called egg gourd was often described separately from other orna- 
mental gourds. While still advertised as an ornamental, this particular variety was 
marketed primarily as an egg mimic. As home production of eggs waned, the egg 
gourd fell out of favor. By the 1930s pepo gourd cultivars were advertised for their 



ornaments 



How does this review strengthen or weaken the "recent escape'' hypothesis 
advanced by the Asches and others? Ornamental Cucurbita pepo gourds have been 
commercially available for nearly 200 years; seven different varieties have been 
marketed since the 1870s. Clearly, there have been opportunities tor all to escape 



32 



COWAN & SMITH Vol. 13, No. 1 



predominate 



Midwest 



egg gourd is the more common, though a green and white striped form also 
occurs. If these free-ranging populations represent escapes, why didn't other 
varieties do the same? Why don't warty, crown of thorns, orange, bi-color, and 
spoon varieties also occur outside of cultivation? Our seed catalogue survey sug- 



more 



marketed through the first half of the nineteenth century. We believe the absence 
of these fancy gourds in natural settings is significant evidence arguing against the 
possibility that Cucurbita pepo gourds frequently "escape" from cultivation. Exam- 
ining the niche and preferred habitat of free-ranging egg and green-striped 
gourds provides further evidence relative to this issue. 



THE NICHE AND HABITAT OF FREE-RANGING GOURDS 



Few detailed descriptions of the niche and habitat of Cucurbita pepo ssp. ovifera 
var. texana and related free-living populations are available. Most habitat informa- 
tion regarding these plants consists of brief field observations, often in the form of 
locational descriptions on herbarium sheets. To augment these records, we stud- 
ied gourd populations in the field. The following observations are based upon 
studies of free-ranging gourd populations in the Missouri and Arkansas Ozarks in 
November, 1990, and in Western Kentucky during November, 1991. 

Efforts were concentrated in the Missouri and Arkansas Ozarks since (1) this 
area was noted by Steyermark (1963) as containing county records of a free-ranging 
cucurbit; (2) a large portion of the Ozark Plateau is densely wooded and sparsely 
populated; and (3) a number of river headwaters are located in the Ozarks, and if 
cucurbits were located in their upper reaches, they might be found in more dis- 
tant, downstream areas. 

Studies in western Kentucky were concentrated in Union County, in the Ohio 
River floodplain several miles below the confluence of the Ohio and Wabash rivers. 
Union County was chosen because of the senior author's strong consanguineal 
ties and familiarity with the area. Cowan was afforded access to property and local 
informants that proved invaluable in the search for free-ranging gourds. 



The Ozarks. 



Missouri 



the White and Buffalo Rivers in Arkansas. In addition, collections made by 
Hoffman and students in the western Ozarks were studied. Descriptions of these 
collections have been documented elsewhere (Smith et ah 1992). The Buffalo River 
populations, however, provide a typical example of the types of habitats free- 



measure 



cess outside modern agroecosy stems. 



com pa 



terms 



removed 



ley, and the Ozarks in general, also represent an excellent potential heartland and 
refuge area where a wild indigenous gourd could have grown in relative obscurity 
until Steyermark collected it in the 1950s. 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 33 



For the past 20 years the entire main valley corridor of the Buffalo river (388 
square kilometers) has been almost free of human habitation, much less agri- 
cultural fields. While there are small family farms along the tributaries of the 
Buffalo River today, the river's watershed of 555 square kilometers is heavily for- 
ested and remains one of the least populated regions in Arkansas, with Newton 
County averaging 18 persons per square kilometer (Pitcaithley 1987:89). The his- 
tory of human occupation of the Buffalo River valley has been well documented 
(Pithcaithley 1987), and provides a picture of sparse population and limited agri- 
cultural development. The area did not witness a post-World War II boom in 
soybean cultivation, and throughout its history would have provided few oppor- 
tunities for survival of a narrow agricultural niche pepo weed. Because of its rela- 
tive isolation and limited human occupation, the Buffalo River valley does, on the 
other hand, provide an opportunity to establish the presence (perhaps long-term) 
of populations of an indigenous wild gourd far removed from agricultural settings. 

The Buffalo River valley also provides an opportunity to consider the general 
habitat requirements of the free-living Cucurbita pepo gourd. The town of Pruitt 
marks a transition point for the river as it widens from a narrow rocky valley floor 
with swiftly flowing water into a broader floodplain having reduced water velocity 
and a meandering channel with associated sand and gravel bank, bar, and island 

formations. Extensivp snrvpv ahnvp Pruitt vipldpd nn pvidpnrp of thp nlant whilp 



common 



areas surveyed below the Nars (Fig. 2). Gourd populations were noted in diverse 
settings within these general habitats. Typically, plants were rooted on the floor of 



dominated 



4m 



less shaded areas where they assumed a procumbent habit or overtook herba- 
ceous annuals such as ragweed and cockleburr. 

The populations encountered in the Ozarks also provided information on 



mechanisms 



mammals 



ing dispersal was largely, but not completely, accomplished by other means. Ample 
evidence of dispersal of dried, buoyant gourds by spring and winter floodwaters, 
however, was noted. All phases of the floating gourd dispersal process were 
observed along each of the Ozark stream and river courses we surveyed. We 
observed gourds resting well within the reach of spring floodwaters. In at least 
one location last year's gourd crop was also observed in transit, floating within an 
abandoned navigational lock chamber at Batesville, Arkansas. We also saw gourds 
in a wide variety of locations where they had been carried by floodwaters, such as 
caught against bushes or other floodwater "filters," trapped inside hollow stumps, 
lying on the ground where they were left by receding floodwaters, deposited 
along roadside ditches and bridge approaches adjacent to river and stream val- 
leys, and frequently in piles of wood and other debris deposited by powerful 



flood 



s. 



This last depositional context, which often places the gourds in juxtaposition 
with objects discarded by humans, is likely the source of occasional observations 
that this plant grows in "dumps" and "waste places," and the conclusion that it 
therefore likely represents an escape from cultivation. The river or stream valley 



34 



COWAN & SMITH 



Vol. 13, No. 1 




H K 



1 

OJ 

> 

G 

2 

c 



Cfl 



X 

0> 



o 

C 

XJ 



en 

g 

73 



1 

en 



03 

y 

o 






C 



en 

0) 



C 

H OJ 



> 



u 



*J3 



9 

CO 

CLI 

X> 



O 

u 

CD 

CD 

en 

O 



O 

c 

o 



en 

CD 



<D 



OS 



o 

en 
OJ 



O 

U 
I. 






<J 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 35 



location and taphonomic context of such deposits identifies them as an inherent 
component of floodplain environments, rather than directly resulting from human 
disposal activities. Bringing the process of seed dispersal full circle, vines of plants 
in a number of locations originated in the fragmented remains of dry brown gourds 
deposited by flood waters of the previous spring. 



Western Kentucky.— In Union and Henderson Counties, Kentucky free-living Cucur- 
bita populations, known as "Johnny Gourds," have been growing for years. Most 
farmers indicated they have always been present in the floodplain of the Ohio 
River, but have become a much greater problem with the draining and opening up 
of formerly forested floodplain areas. 

Typically, "Johnny Gourds" emerge in the spring from the forest lining the 
bank of the Ohio, or from drainage ditches far removed from the river. As the 
growing season progresses, the vines aggressively invade fields where, unless 
controlled, they overtake corn and soybean crops. They can be checked by pre- 
emergent and broadleaf herbicides, but are virtually impossible to eradicate. Col- 
lections made near DeKoven, Kentucky, provide a good example of this problem: 
a drainage ditch had been sprayed at least twice but we collected dozens of gourds 
from both the bottom of the ditch and the field margin. 

This cultivated field habitat is far different than that we encountered during 
our Ozark survey, and accounts for the origin of the "recent escape" hypothesis. 
In addition, all Union County fruits showed evidence of introgression with other 
cultivars. Both the Asches (1992) and Wilson (1990) made similar observations 
about collections from a modern agroecosystem on Kaskaskia Island, Illinois. On 
the other hand, local informants suggest that while "Johnny Gourd" populations 
have become more abundant in the past 30 to 40 years, they have been present far 
longer. The fact that these western Kentucky populations exhibit extensive intro- 
gression with cultivar pepo varieties is the logical outcome of their frequent exposure 
to garden crops along the populated Ohio River. 



MORPHOLOGICAL CHARACTERISTICS OF MODERN 
Cucurbita GOURDS AND THEIR RELATIONSHIP 

TO ARCHAEOLOGICAL CUCURBITS 



information from 



provides strong support for the argument that contemporary eastern Cucurbita 
pepo gourds are not simply "escapes" from cultivation or recently introduced agri- 
cultural weeds, but are rather part of an indigenous flora. Viewed in this light, the 
wild gourds are the closest, and most logical ancestors for the eastern lineage of C 



Walters 



me 



if 



acteristics of these populations, and compare them with their cultivated counter- 



from _ 
Morphological informat 



ing sources: (1) a sample of wild morphotype fruits collected from the Ozarks in 
1990, (2) a sample of cultivated varieties of C. pepo ssp. ovifera grown in a field 



36 



COWAN & SMITH Vol. 13, No. 1 



) samples or cultivated 
Wilson (1986). Metric 



sam 



Newsom of the Florida Museum of Natural History, while that from 
Missouri was provided bv Smith. 



measurements 



tch from i 
diameter 



were measured (Fig. 3, Table 2-3). Statistical information from other contempo 
rary and archaeological Cucurbita populations are listed Tables 4-5. 



Fruit size, shape, and color.— Fruits from our Ozark sample are small and exhibit a 
limited number of shapes (Fig. 4). Maximum height ranges from 3.9-10.0 cm 
(mean 5.9 cm) (Table 2). Maximum fruit diameter is less variable than height and 
averages 4.6 cm (range 3.2-5.5 cm). Fruit shape varies from nearly circular, to 
oblate, to pear-shaped with most assuming a slightly prolate form (Fig. 4). This 
geometry can be expressed as a ratio which incorporates both maximum fruit 
height and maximum fruit diameter where ¥ shape = maxht/maxdia. In this ex- 
pression, ¥shape is a measurement of the circularity of the fruit; in a circular fruit 
this ratio is one. The closer the ratio is to one, the more globular the fruit. As can be 
seen in Fig. 5, this objective classification yields results which confirm our subjec- 
tive observations: most fruits are slightly prolate. The two specimens greatly 
exceeding one are classified as pear-shaped. 

Two distinct fruit colors were noted in the Ozark sample. By far the majority were 
ivory; immature fruits of this type were pale green. A small number of fruits 
were pale green to ivory with dark green stripes. When stripes occurred, they 
were invariably 10 in number. Striped and unstriped fruits were never noted 
growing on the same plant, although plants producing the two fruit colors were 
found on the same sand or gravel bar. 

The epidermis of the fruits was almost always smooth and unlobed. Although 
a few fruits did possess minor epidermal warts, these were infrequent, and seemed 
to have developed as the result of some injury to the fruit wall as it was maturing. 
The lack of wartiness and lobing is significant for it suggests these gourds have not 
experienced extensive introgression with cultivars. We specifically noted a lack of 
introgression with two common edible cultivars— the yellow crookneck and the 
acorn squash. Crookneck cultivars are often strongly warty and yellow in color; 
acorn cultivars are invariably strongly lobed and dark green. Warty ornamentals 
of the yellow-flowered gourd (C. pepo ssp. ovifera) might also be expected to be 
grown in the Ozarks, but no evidence of their contaminating presence was noted 
in our collections. In the two instances of introgression we did note, fruit shape 
and coloration suggest hybridization with other Cucurbita cultivars. One of these, 
collected from a driftwood pile on the Gasconade River in Wright County, Mis- 
souri, seems suggestive of crossing with butternut (C. moschata); genetic analy- 
sis (Deena Decker- Walters, personal communication, 1992; see Smith et al. 1992, 
Fig. 4.12) supports this. In addition, two fruits were collected from a plant in the 
White River valley of Searcy county, Arkansas, that were yellow-orange with pale 
yellow stripes. 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



37 



u 



G 






FIG. 3.— Schematic 



it 



maximu 



knob 



tf 



thickness; D: corolla, or blossom, "knob" thickness; E: thickness adjacent to co- 
rolla "knob"; F: thickness adjacent to peduncle "knob"; G: thickness at maximum 
fruit diameter. 



38 



COWAN & SMITH 



Vol. 13, No. 1 




























65-1 








5 



10 



I — I — I — I I I «t»»i 



cm 



FIG. 4.— Fruit shapes for selected Ozark cucurbit fruits. Numbers correspond to 
field specimens. 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 39 



TABLE 2.— Measurements of fruit size, shape, and rind thicknesses for selected 
Ozark cucurbits. 



Field/ Max Max Thk Thk Thk Ped Bios 

Lab No. 1 Hght 2 Dia Fshape 12 3 Thk Thk 

3-1 6.1 5.1 1.19 1.6 1.3 1.5 4.9 4.9 

4-1 8.3 4.7 1.76 1.8 1.5 1.9 3.8 4.4 

5-1 4.0 5.2 0.76 0.9 0.8 0.9 4.0 3.7 

9-2 5.2 4.6 1.13 2.0 1.1 1.4 5.6 3.2 

9-4 5.3 5.3 1.00 2.0 2.0 2.0 5.8 6.0 

11-1 4.2 4.0 1.05 1.4 1.1 1.2 4.9 5.0 

14-1 6.0 5.0 1.20 1.6 1.4 1.9 4.7 4.4 

14-3 4.3 4.9 0.87 1.6 1.2 1.6 5.1 3.2 

14-5 5.3 4.1 1.29 1.9 1.2 1.9 4.7 4.6 

14-6 6.1 4.7 1.29 1.9 1.2 2.2 4.5 6.2 

14-9 3.9 3.2 1.21 2.0 1.1 1.4 5.6 3.2 

14-10 8.4 5.2 1.61 1.5 1.0 1.3 3.5 6.2 

14-11 5.7 4.5 1.26 1.8 1.9 1.9 7.7 6.0 

64-2 6.1 5.5 1.10 1.5 1.0 1.4 5.0 4.8 

65-1 10.0 4.4 2.27 1.5 1.3 1.3 4.8 3.8 



^ey to Field/Lab No: 3-1 Wilbur Allen Wildlife Refuge, Wright County, Missouri; 4-1 Hodgson Mill, 
Ozark County, Missouri; 5-1 Batesville, Independence County, Arkansas; 9-2, 4 Highway 9 Bridge, 
Woodruff County, Arkansas; 11-1 Manes, Wright County, Missouri; 14-1,3,5,6,9,10,11 Highway 14 
Bridge, Searcy County, Arkansas; 64-2 Highway 64 Bridge, Woodruff County, Arkansas; 65-1 High- 
way 65 Bridge, Searcy County, Arkansas. 

2 See Figure 3 for these measurements. MaxHght = maximum height; MaxDia = maximum diameter; 
Fshape = fruit shape (MaxHght/MaxDia); Thkl-3 = rind thickness 1-3; PedThk = peduncle thick- 
ness; BlosThk = Thickness of rind at blossom. All measurements in cm. 



Rind thickness.— It has been argued that rind thickness is an important factor in 



rrninin 



America. Smith (1987) argues that archaeological squash rind less than 2 mn 
maximum thickness and dating before 3000 B.P. can not be used as evidence 
domesticated forms, since rind of this thickness falls within the size range of v 
morphotype gourds. Our Ozark collections provide additional information 
this point. 



measured 



Table 2). Rinds of all fruits were thicker at the peduncle and blossom (corolla) ends, 
where a corky "knob" often forms in the fruit cavity. Thickr ' " l 



"knobs" 



the 



mm (ranee 3.2-6.2 mm) at the blossom end. With 



(Table 2). Rind is thinnest at the point of greatest maximum deflection. Variability 
in thickness, as expressed by the coefficient of variation (CV) (Simpson et al. 
I960), is quite low, suggesting that this morphological characteristic is conser- 
vative, and the Ozark population fairly homogeneous (Table 2). 



forms of Cucurlnta pq?o 



40 



COWAN & SMITH 



Vol. 13, No. 1 




OS 




o 



o 




Li. 



o 



Q 




N.* *v <V *v *v Oy fc # N* *b* ^>* fc* 9>* 

•T tr «r N v # qr qr N tr N v* N w° N tf° ^ 



Field Number 

FIG. 5.— Bar graph of Ozark fruit shapes. Note the number of fruits falling close to 
1, indicating a relatively circular shape. Spikes indicate pear-shaped fruits. 

quite variable. Differences in rind thickness among pumpkins, acorns, and croo - 
necks seems to be related to variation in wall thickness associated with lobmg an 



forms invariablv have a non-textured epidermis 



uniform 



mm 



domesticated 



domestication 



accompanied by variability in epidermal 



ding lobing and wartiness (Cowan 1990). The well-preserved pre-4000 B.P. find 



mm 



textured, with no evidence of lobing or wartiness, suggesting that they are not t 



domesticated form 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



41 



TABLE 3.— Measurements for Ozark gourd seeds. 



Field/No. 1 



3-1 

4-1 

5-2 

9-2 

9-4 

11-1 

14-1 

14-3 

14-5 

14-6 

14-9 

14-10 

14-11 

64-2 

65-1 

Sample Mean 



Mean 
Lngth 2 



9.8 
9.1 
9.3 
9.0 
9.2 
9.4 
9.1 
9.4 

9.5 
8.2 
7.8 
9.2 
8.7 
10.1 
9.2 
9.2 



Range 



Std. 
Dev. 



9.0-10.6 
7.9- 9.6 
8.6-10.0 
8.3- 9.4 
9.0- 9.9 
8.9- 9.9 

in- 9.8 

8.7-10.2 
8.2-10.2 
7.6- 8.6 
7.0- 8.3 
8.5- 9.5 
8.3- 9.0 
9.5-10.9 

8.2- 9.9 
Sample Mean 



'See Table 2 for key to field numbers. 

2 MeanLngth = mean length; Std. Dev. 
surements in mm. 



.43 

.52 
.41 
.34 
.23 
.29 
.45 
.41 
.49 
.24 
.34 
.25 
.21 
.40 
.46 



Mean 
Wdth 



6.4 
5.8 
6.0 
5.7 
5.9 
6.7 
5.9 

6.1 
6.4 

4.9 

5.2 

5.7 

5.6 

6.5 

5.9 

5.9 



Range 



6.0-6.9 

4.9-5.8 

5.7-6.4 

5.3-6.0 

5.6-6.3 

6.6-7.0 

5.3-6.5 

5.8-6.4 

6.0-7.1 

4.7-5.2 

4.8-5.7 

5.5-6.0 

5.2-6.0 

6.2-7.0 

5.6-6.3 



Std. 
Dev. 



.26 
.43 
.20 
.19 
.19 
.14 
.27 
.17 
.37 
.13 
.24 
.14 
.20 
.18 
.18 



tandard deviation; MeanWdth = mean width. All mea 



The 2 mm maximum thickness Dasenne we * 
rly domesticated forms works best when sam 
ment greater than 2 mm can probably be safely 
?ss it comes from the blossom or peduncle port 



lm 



domesticated 



mm are eood markers 



sim 



Seed size and characteristics. -A sample of 20 seeds was selected from each fruit 



Mean 



(range 7.8-10 1 mm); mean width 5.9 mm (range 4.9-6.7 mm). The coefficient of 
variation were quite low, suggesting a fair degree of conservatism in seed 
size within the population. Not surprisingly there is a P osltlv .\ rel ^° n . S h h, P 
between seed length and width (r = 0.85). Seed size (measured by length x 
width), however appears unrelated to fruit size (measured by height 



width), however, appears 
diameter) (Fig. 6). 
Compa 



pre-4000 B.P. (Cloudsplitter, Kentucky and Phillips Spring, ™" ™" '^ 
prehistoric (Hontoon fs.and, Florida) ««, ;, *e ■*"£«££»%. 



commerci 



When 



I table 4, hie. 7). When arcnaeoiogicai wu. F .« — — - r standard 

tions, two factors stand out. First, there is considerable overlap in the standard 



42 



COWAN & SMITH 



Vol. 13, No. 1 




■ ■■ 

X 




c 







N 

CO 





CO 



80 



70 




60 






50 



40 




30 









¥ 





10 



20 



30 



40 



50 



Fruit Size (height x diameter) 



FIG. 6.— Bivariate plot of the relationship between seed size (length x width) and 
fruit size (fruit height x fruit diameter). The random scatter of the points suggests 
little relationship between seed and fruit size. 



deviations of the lengths and widths of the seeds of these populations (e.g., Fig- /)• 



most 



ticeable in the cultivars and in the archaeological population from Phillips Spring. 



measure 



popu 



intermediate position between most 



vifera vai 
modern 



-■/< 



possess larger coefficients of variation. In part the large CV value for modern 
cultivars is the result of pooling varieties with radically different fruit mor- 



phologies 



from Phillips Spring m 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 43 



TABLE 4.— Summary statistics for various Cucurbita seed populations. 



Mean Mean 

Population N 1 Length CV Width CV 



Ozarks 300 9.2 5.4 5.9 6.6 

Texana 60 9.6 2.1 6.2 3.7 

Cultivars 70 9.3 9.0 5.8 6.8 

Cloudsplitter 2 8.7 1.6 5.4 5.1 

Phillips Spr. 2 49 10.2 8.7 6.9 9.1 

Hontoonls-SN 149 8.8 3.5 5.8 5.3 

Hontoonls-SH 868 9.0 2.2 6.0 1.5 

Hontoon Is-1 1,094 8.9 3.0 5.9 3.8 



T N = numbers of seeds measured; CV = coefficient of variation; Ozarks ■ pooled sample of seeds 
from 15 fruits; Cultivars = pooled sample of C. pepo ssp. ovifera reported by Decker and Wilson 1986; 
Texana = C. pepo ssp. ovifera var. texana reported by Decker and Wilson 1986; Cloudsplitter = Cloud- 
splitter shelter, Kentucky, seeds measured by Cowan; Phillips Spr. = Phillips Spring, Missouri, seeds 
measured by Smith; Hontoon Island-SN - "Snail Midden" complete seeds; Hontoon Is-SH = 
"Mussel Midden" complete seeds, Hontoon Is-1 = combined proveniences, including a shell-free 
midden. Hontoon Island measurements provided by Lee Newsom. 

2 In her landmark study, King (1985) had a larger collection of seeds available for study than was 
considered here. As a result, she obtained slightly larger mean length (10.5 mm) and mean width (70) 
values from a sample of 65 Phillips Spring seeds. 



more 
simili 



Based upon these data, seed size alone does not allow one to distinguish the 



pepo 



While 



of cultivated ornamentals seems more variable (as expressed by standard devia- 
tion and the coefficient of variation), this is hardly a useful measurement if dealing 
with archaeological collections of only one or a few seeds. Again, while it is possi- 



mm 



ticates (Smith 1992:45), a single s< 
from either a wild or domesticated 
direct date olav an important role i 



Seed and rind bitterness.— Chemical compounds in both the rind and flesh of truly 



im 



serve 



render the fruits virtually inedible to humans. The fact that all cultivated forms of 



f 



im 



With tl 
mesticated 



200) 



that were usually bitter tasting because placental tissue adhered tightly to the seed 



44 



COWAN & SMITH 



Vol. 13, No. 1 



E 

E 




o 
c\i 



o 
o 

o 










o 

o 

CO 




o 




CD 



O 

o 











CD 




c 



CO 





CD 



in 



C/> 




o 

o 



a. 

CO 

co 

a. 



CO 






CO 



o 
o 




o 
o 




o 
o 




CO 
N 

o 




CO 



CO 



8 





o 




c 

o 





o 



0- 




(0 

> 



o 






a 

C/) 
CO 




g 

co 




O) 





CD 



0) 






CO 



CO 
CD 

E 








co 



co 2 

CO 




o 




CO 




"a 




CD _ 

"O CO 

CD 






CO 

CO 




o 

CO 



CD 

c 

o 



3 



CD 

t3 




o 

> 

o 



o 

or 
aj 

WD 



c 

CD 

c 

o 



A3 




> 

a» 

c 

03 



CD 

c 

o 



QJ 



O 



73 



C 

A3 



bJD 

a; 
a; 

CD 

C 



d 



y- 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



45 





o 




LU 




Q 

LU 

LU 

CO 




o 



10 



8 






T 



I 



T 



T 




CULT 




PHSP 





OZARKS 



4 



HSN 



HON1 





2 







HSH 




TEX 




CLDSP 




i 



I 



1 



1 








2 



4 



6 



8 



10 



12 



CV SEED WIDTH 



FIG. 8.— Bivariate plot of the relationship between coefficients of variation of seed 
length and seed width. Key to symbols: HSH (Hontoon Island shell layer), Honl 
(Hontoon Island), HSN (Hontoon Island snail layer). All measurements supplied 
by Lee Newsom; TEX (Cucurbita pepo ssp. ovifera var. texana, Decker and Wilson 
1986, Table 3); CLDSP (Cloudsplitter Shelter, KY; measured by Cowan); OZARKS 



M 



MO; measured 
ifera, Decker and Wilson 



// 



nest egg," "white egg, 



/' 



// 



flat striped," "bicolor pear," "striped pear," and "white pear"). 



number 



coat. Production of a large number ot bitter tasting seeus is tunua.j w .... , «~.~- 
characterization of this plant as lacking these attributes of a successful pioneering 
species (Asch and Asch 1992), and is consistent with a successful adaption to 
natural floodplain environments. Of the fruits collected during our survey, and 
those sent to us by other collectors, only a handful produced nonbitter seeds. 
Interestingly, field observations at the time these fruits were collected led us to 



46 



COWAN & SMITH Vol. 13, No. 1 



overwhelmin 



important com 
ilations. When 



between wild and cultivated pepo squashes occurs, bitterness is lost. The lack of 



"sweetness" 



material 



domesticated and wild squashes. 



lm 



human foraeers. While it is virtually impossible to remove 



min is sufficient to render them 



sim 



economically im 



millennium 



As we have noted elsewhere, large "patches" of gourds can produce pro- 
digious quantities of fruits. For example, a single patch on the Buffalo River cover- 
ing about 200 square meters produced over 100 fruits containing 10,000-20,000 
seeds (Smith et al. 1992:94). Under cultivated conditions, wild gourds could be 
expected to produce even greater quantities. Preliminary analysis of the nutri- 
tional qualities of wild gourd seeds suggests that more than 24% of their dry 

1QQ9V thiQ pniials the nrotein content of 



Smith 



domesticated sunflower 



1992, Table 9.3). Smith's analysis assayed the content of the entire seed (including 



mi 



mi 



Peduncle absciscence and diameter.— Whenever possible, peduncles were collected 
with each fruit, but most fruits had already abscised from the vines. In general, 
this seemed to be a function of latitude and altitude; more southerly, lower eleva- 
tion collections were generally less mature, and fruits more likely to be retained 
on the plant. Even when fruits were found on the vine, however, it was difficult to 
remove both the fruit and peduncle from the vine without having the fruit detach 
from the peduncle. Fruits often abscised from the peduncle with the slightest tug. 
Andres (1987) has noted this peculiarity in var. texana, and suggests it represents 
an adaptive strategy that allows fruits to be readily dispersed during floods. 

Recent collections of C. pepo ssp. ovifera and C. pepo suggest that absciscence 
among cultivated ornamental gourds and squashes and pumpkins is highly vari- 
able. Larger fruited forms of cultivated C. pepo (acorn squash, Connecticut field 
pumpkin) remain firmly attached to the vine long after fruits mature. On the other 
hand, absciscence in ornamental gourds varies from one variety to the next. In 
recent collections of mixed ornamentals, the senior author noted that forms such 
as "Green-striped pear" were most susceptible to falling free of the vine. Warty 
fruits were the most persistent; rarely could a fruit be collected without breaking 
the peduncle from the vine. Green striped, oblate, or cheese-shaped forms were 
intermediary between the "pear" forms and the "warty" ones. 

Peduncles on Ozark gourds are uniformly five-angled, a common character of all 
varieties of C. pepo. In a sample of 20 fruits and associated peduncles, diameters (max- 
imum distance between two points at the peduncle base) ranged from 5.3-8.6 mm, 
with a mean of 6.6 mm (standard deviation 0.79 mm) (Table 5). This measurement 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 47 



TABLE 5.— Peduncle diameter and fruit size in free-living Ozark and cultivated 
Ohio Cucurbita gourds. 



Peduncle Maximum Fruit Maximum Fruit 

Diameter (mm) Height (cm) Diameter (cm) 



Ozark Sample 1 

33 
39 

40 
42 
44 

48 

123 
124 

125 
133 
166 
167 
168 
285 
286 
288 
301 
309 
389 
399 



6.1 5.5 4.7 

6.7 5.6 4.8 
5.6 5.0 4.2 

6.5 5.4 4.8 
5.9 5.6 4.5 

6.2 5.5 4.8 

6.6 8.9 5.2 

5.3 7.6 4.5 
5.9 5.2 4.6 
6.3 5.7 5.3 
7.6 5.0 5.0 
7.6 5.3 4.5 

7.3 4.6 4.5 

6.4 5.7 4.7 
6.3 5.1 4.4 

5.8 4.0 3.9 
7.3 4.9 5.4 
6.6 5.0 5.2 
8.6 6.6 6.7 

6.9 6.0 5.8 



Sample Mean - 6.6 Std. Dev. .80 



Cultivar Sample 2 

1 

1 
1 

1 
2 

2 
2 
3 
3 
3 
3 
3 

4 
4 
4 
4 

4 
4 
4 
4 



8.5 4.7 8.5 

8.3 4.8 8.0 
10.0 4.5 7.7 
12.4 4.7 8.6 

7.5 4.2 4.3 

6.4 4.9 4.5 
6.2 4.2 3.9 
8.0 6.2 8.8 
7 6.4 8.7 
7.9 6.1 9.4 
7.0 6.6 9.3 
9.0 4.5 7.5 

13.7 7.2 7.1 

12.0 7.3 7.4 

15.0 11-5 85 

11.0 117 6.9 

10.0 6.3 6.8 

10.0 6.4 7.1 

15.1 8.3 9.2 
14.9 8.2 8.2 



48 



COWAN & SMITH Vol. 13, No. 1 



TABLE 5.— Peduncle diameter and 
Ohio Cucurbita gourds, (continued) 



Peduncle Maximum-Fruit Maximum Fruit 

Diameter (mm) Height (cm) Diameter (cm) 



■— ^^ 



Cultivar Sample 

4 

4 

4 

4 



13.0 8.0 8.2 

11.3 7.1 7.2 

10.0 6.4 7.2 

12.9 5.2 8.7 



Sample Mean = 10.3 Std. Dev. 2.9 



•Ozark Cat. Nos. 33-48 Spavinaw Creek, Benton County, AK; 123-133 Wilbur Allen Wildlife Refuge, 
Gasconade River, Wright County, MO; 166-168 Maumee Landing, Buffalo River, Searcy County, AK; 
285-286 Highway 14 Bridge, Buffalo River, Searcy County, AK; 288-309 Highway 9 Bridge, White 
River, Izard County, AK; 389-399 Red River, Bossier Parish, LA. 

^Cultivars (all from field of mixed ornamental C. pepo ssp. ovifera growing in Hamilton County, Ohio)^ 
1 Green and white striped, with shallow lobes, oblate shaped; 2 yellow-striped, slightly warty and 
lobed; 3 dark green stripes on light green background, slighted lobed and slightly warty, oblate 
shaped; 4 yellow warted, oblate to pyriform shaped. 

was significantly larger (mean 9.75 mm, range 6.2-15.1 mm, standard deviation 
2.8 mm) in a sample of 23 peduncles of mixed cultivated ovifera fruits. 

Peduncle diameter in Cucurbita pepo is clearly related to fruit volume. This is 
visually obvious when comparing the large fruits of some cultivated pepos, but is 
less so for cultivated ornamentals and free-ranging gourds. A comparison of indi- 
vidual peduncle diameters from cultivated ornamentals from southwest Ohio and 
free-ranging Ozark fruits serves as an objective measure of this relationship, an 
clarifies the utility of the peduncle in determining wild versus cultivated status ot 
small pepo forms. As illustrated in Fig. 9, most free-ranging gourds produce pe- 
duncles and fruits that are smaller than cultivated C. pepo ssp. ovifera. While there 
is overlap in the size distributions, we believe peduncle diameter is proba y 
useful for determining whether archaeological specimens are cultivated or wil 
one has a collection of sufficient size. We suggest, for example, that in gener , 
peduncles with diameters of 8 mm or less are probably from wild fruits, and tnos 
with diameters larger than this are likely domesticated. Since peduncle diameters 
of wild and cultivated forms overlap, however, this cut-off is best used with cau- 
tion. And, like rind and seed measurements, archaeological sample size limits 
utility of any peduncle diameter "index" in distinguishing between wild and early 
domesticated forms. 

SUMMARY OF MORPHOLOGICAL CHARACTERISTICS AND THEIR 
INTERPRETIVE VALUE FOR ARCHAEOLOGICAL COLLECTIONS 



Morpholo 



information useful in evaluating wild versus u 
mH arrhapnlnairal specimens Tn oarticular, these 



suggest that the following characteristics can be used as markers of domes 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



49 



CD 



CD 

E 

CO 

b 

o 





CD 
Q_ 



16 



14 



12 



10 



8 



6 



4 





o 



15 



CO 

o o 





o 




o 



o 






• 






o 



o 



o 



o 



o 








• 









32 



49 



66 



83 



100 



Fruit Size (height x diameter) 




Cultivated ornamental gourd 



o=W 



FIG. 9.-Bivariate plot of the relationship between peduncle diameter and fruit 



maximum heieht x maximum 



population grown in Hamilton County, Ohio. 



</< 



(1) rind thickness which exceeds 2 mm; 

(2) epidermal lobing and/or warting; 

(3) seed length greater than 11 mm; and 

(4) peduncle diameter exceeding 8 mm. 



These guidelines for identifying wild versus domesticated pepo do little to alter 
the current model of the appearance of truly domesticated Cucurbita pepo in east- 

_ *• - rr . ^y . ... n A A\ Affor An unknown 



America (see Smith 



similar 



50 



COWAN & SMITH Vol. 13, No. 1 



Ozarks today, domestication occurred sometime between about 4300-3000 B.P. 
The Phillips Spring assemblage (ca. 4300 B.P.) is important in this regard, in that it 
might represent an early stage of domestication. Length and width of Phillips 
Spring gourd seeds, and their range of variation, fall on the edge, or outside the 
range of wild forms, while peduncle diameter and rind thickness fall well below 
the boundary lines of domestication. 



FREE-LIVING Cucurbita pepo GOURDS: SUMMARY 



information from herbarium 



most 



habitat of free-living Cucurbita pepo gourds can be offered. 



America 



stream 



stream-s 
ts match 



buoyant gourds are often trapped by weeds, bushes, and other floodwater "fil- 



ters," or are deposited with other floodborne materials in debris piles. Whether 
the context of deposition is stream-side sand and gravel bars and hummocks, as 
documented along the smaller streams and rivers of the Ozarks, or the higher 
elevation sandy terraces and levee ridges of larger river valleys, gourds colonize 
open edge areas within the constantly reworked floodplain landscape. These sunny 
to partly shaded habitats constitute the annually disturbed habitat setting within 
which Cucurbita pepo gourds are strong and successful competitors. With rapidly 



more than 30 meters 



lm 



blossoms 



sometimes 



With 



pollination, isolated plants are not disadvantaged by wide floodwater dispersal. 
Contrary to the Asches' assertion, there is no evidence that var. texana (or any of 
the free-living Cucurbita pepo gourds) do not have the ability to self-pollinate (Deena 
Decker-Walters, personal communication, 1992). 

These adaptational aspects of free-living gourds— buoyant, hard-walled fruit 
functioning for efficient seed dispersal by floodwaters, aggressive climbing and 
growth characteristics of vines, prolific gourd and seed production, and self- 
pollination-all suggest long-term evolution within and adantion to. river flood- 



environments. This 



nd 



maintained floodplain habitats such as bridge approaches, drainage ditches, I 
agricultural fields. This plant's well documented ability to infest fields within the 
reach of floodwaters chronically certainly qualifies it as a floodplain agricultural 
weed of the first rank. But as indicated by the above habitat descriptions, it is a 
mistake to characterize the free-living gourds narrowly as agricultural weeds, and 
certainly in error to view them as closely associated with soybean cultivation. On 
the contrary, it is important to begin any characterization of free-living Cucurbita 
pepo gourds with a recognition of their successful adaptation to naturally dis- 



Summer 1993 JOURNAL OF ETHNOBIOLOG Y 5 1 



turbed floodplain environments, and to then view their success in anthropogenic 
contexts as the simple expansion of the plant into areas where human activities 
have expanded, sometimes dramatically, the habitat of the plant. 

Based upon information collected from a survey of archaeological literature, 
herbaria holdings, and field and laboratory observations, we have been able to 
document the following: 

(1) Archaeological evidence of a morphologically wild Cucurbita pepo gourd is 
present in broad areas of the East spanning a period of nearly 12,000 years. 

(2) Historical evidence of this same plant can be traced from the discovery of 
Cucurbita pepo ssp. ovifera var. texana in 1835, and throughout the remainder of 
the nineteenth and early twentieth centuries in the central Mississippi valley 
where a wild morphotype gourd has been repeatedly collected since 1846. 



Southeast 



Midw 



(4) Rather than narrow niche agricultural companion weeds, free-ranging gourds 
are well adapted members of riparian plant communities and thrive in habitats 
far removed from agricultural settings. Although the same gourds do invade 
cultivated fields, they are by no means tied to such human-maintained habitats. 

While ongoing analyses of archaeological materials and contemporary collec- 
tions will undoubtedly flesh out the relationships between cultivated forms of C. 

ra and free-living Cucurbita pepo gourds of eastern North America, 
~w ^^^ mere is little need to rely on Mexico as the "hearth" from which cucur- 
bits diffused into eastern North America. Indeed, these data call into question the 



<y< 



Mexico 



from the central and northern Mexican 



radiometrically 



am 



We 



companion article by Decker- Walters 



issue provides further, and we think, compelling support for this position. 



NOTES 



ih 



We 



pepo ssp. oviji 



volume). 



* 



2 The following seed catalogues are cited in the text. All are on tile at the United States 
Department of Agriculture, National Agricultural Library in Beltsvdle, Maryland. 

BLISS, B.K. 1860. A Descriptive Catalogue Containing a Choice Collection of Flower Vege- 
table, and Agricultural Seeds. Eighth annual edition. Springfield Massachusetts. 



FERRY 



Wooster 



uie etna r lower seeas. vvoosier, wmu um ^»-" . v} w , orc „ j 

HOVEY and COMPANY. 1845. Descriptive Catalogue of a Choice Collection ot Rowers and 



Seeds. Boston. 



52 



COWAN & SMITH 



Vol. 13, No. 1 



M'MAHON, BERNARD. 1804. Catalogue of American Seeds. Philadelphia 
1819. The American Gardener's Calendar. Philadelphia. 



O'KEEFE, M. and SONS COMPANY. 1868. Catalogue of Seeds and Guide to the Flower 

and Vegetable Garden. Rochester, New York. 
PRINCE, WILLIAM and SON'S. 1834-35. Annual Catalogue of Esculent Vegetables and 

other Seeds of the Choicest American and Imported Varieties. 
SIBLEY, HIRAM COMPANY. 1879. Catalogue of Garden, Field, and Flower Seeds. Rochester, 

New York and Chicago. 

VICK, JAMES. 1866. Vick's Illustrated Catalogue of Seeds and Guide to the Flower Garden. 
Rochester, New York. 



ACKNOWLEDGEMENTS 

Funds for travel to Missouri and Arkansas were made available through the Research 
Opportunity Fund, National Museum of Natural History, and the Cincinnati Museum of 
Natural History. The following individuals provided the authors with information and wild 
gourds: Louis Baker, Wallace Booth, Dan Dickson, Gayle Fritz, Michael P. Hoffman and his 
students, Cecil Ison, Frank Schaumbach, and Craig and Elizabeth Sheldon. Lee Newsom, 
Florida Museum of Natural History, generously provided measurements from Hontoon 
Island, Florida. Dale Johnson, Martha Riley, and George Yatskievych, Missouri Botanical 
Garden, were extremely helpful in locating information on early St. Louis cucurbit collec- 
tions. Dennis Kearns, University of Texas Herbarium, and Michael Canoso, Harvard Uni- 
versity Herbarium, were similarly helpful in tracking specimens curated at their institu- 
tions. Judith Ho, National Agricultural Library, provided access to that institution's 
extensive collection of nineteenth century seed catalogues. Deena Decker- Walters, Ter- 
rence Walters, Gayle Fritz, Fran King, Lee Newsom, Deborah Pearsall, and Richard Yarnell 
read various manuscript drafts and provided valuable comments and corrections which 
substantially improved the tone and content of the article. Thanks to Al Adamson for 
drafting Figures 3 and 4. Valerie Chaussonnet expertly translated the abstract into French; 
Pilar Smyth performed the same duty for the Spanish version . 



LITERATURE CITED 



ANDRES, THOMAS C. 1987. Cucurbita 
fraterna, the closest wild relative and 
progenitor of Cucurbita pepo. Cucurbit 
Genetic Cooperative Reports 10:69-71 

ASCH, DAVID L. and NANCY B. ASCH. 
1985. Prehistoric plant cultivation in 
west-central Illinois. Pp. 149-203 in 
Prehistoric Food Production in North 
America. Richard I. Ford (editor). An- 
thropological Papers 75, Museum of 
Anthropology. University of Michigan, 



Ann Arbor. 



177 



293 in Geoarchaeology of the Ambrose 
Flick site. Russell Stafford (editor). Re- 
search Series 10, Kampsville Archaeo- 
logical Center, Kampsville, Illinois 
BOYETTE, G., E. TEMPLETON, and L R 
OLIVER. 1984. Texas gourd {Cucurbita 



655. 



649 



CONARD, N., D.L. ASCH, N.B. ASCH, 
D. ELMORE, H.E. GOVE, M. RUBIN, 
J.A. BROWN, M.D WIANT, K.B. 
FARNSWORTH, and T.G. COOK. 
1984. Accelerator radiocarbon dating 
of evidence for prehistoric horticulture 
in Illinois. Nature 308:443-446. 

COWAN, C. WESLEY. 1990. Prehistoric cu- 
curbits from the Cumberland Plateau 
of eastern Kentucky. Paper presented 
at the 47th Annual Meeting of the 
Southeastern Archaeological Confer- 
ence, Mobile, Alabama. 

H. EDWIN JACKSON, 

KATHERINE MOORE, ANDREW 
NICKELHOFF, and TRISTTNE SMARI- 
1981. The Cloudsplitter rockshelter, 
Menifee County, Kentucky: A pre- 
liminary report. Southeastern Ar- 
chaeological Conference Bulletin 
24:60-75. 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



53 



CRITES, GARY D. 1991. Investigations into 
early plant domesticates and food 
production in middle Tennessee: A sta- 
tus report. Tennessee Anthropologist 
16:69-87. 

DECKER, DEENA. 1986. A Biosystematic 
Study of Cucurbita pepo. Unpublished 
Ph.D. Dissertation, Department of Biol- 
ogy, Texas A. & M. University, College 
Station. 

1988. Origin(s), evolution, 

and systematics of Cucurbita pepo (Cu- 
curbitaceae). Economic Botany 42:4-15. 

and LEE NEWSOM. 1988. 

Numerical analysis of archaeological 
Cucurbita seeds from Hontoon Island, 



HEISER, CHARLES B. 1985. Some botani- 
cal considerations of the early domesti- 
cated plants north of Mexico. Pp. 57-72 
in Prehistoric Food Production in North 
America. Richard I. Ford (editor). An- 
thropological Papers 75, Museum of 
Anthropology, University of Michigan, 
Ann Arbor. 

. 1989. Domestication of Cu- 

curbitaceae: Cucurbita and Lagenaria. 
Pp. 472-480 in Foraging and Farming. 
David Harris and Gordon Hillman 
(editors). Unwin Hyman, Boston. 

JONES, F. 1975. Flora of the Texas Coastal 
Bend. Mission Press, Corpus Christi, 
Texas. 



Florida. Journal of Ethnobiology 8:35-44. JONES, GEORGE and GEORGE FULLER. 



DECKER, DEENA and HUGH G. WIL- 
SON. 1986. Numerical analysis of seed 
morphology in Cucurbita pepo. System- 
atic Botany 11:595-607. 

1987. Allozyme variation in 



the Cucurbita pepo complex: C. pepo var. 
oviferavs. C. texana. Systematic Botany 
12:263-273. 

DECKER-WALTERS, DEENA. 1990. Evi- 
dence for multiple domestications of 
Cucurbita pepo. Pp. 96-101 in Biology 
and Utilization of the Cucurbitaceae. 
David M. Bates, Richard W. Robinson, 
and Charles Jeffrey (editors). Cornell 
University Press, Ithaca, New York. 

, TERRENCE W. WALTERS, 
C. WESLEY COWAN, and BRUCE D. 
SMITH. 1993. Isozymic characteriza- 
tion of wild populations of Cucurbita 
pepo. Journal of Ethnobiology 13:55-72. 

EGGERT, HENRY. 1891. Catalogue of the 
Phaenogamus and Vascular Cryptoga- 
mous Plants in the Vicinity of St. Louis, 
Mo. Privately Published. Pamphlet on 
file at the Missouri Botanical Garden, 
St. Louis. 

ERWIN, A.T. 1938. An interesting Texas 
cucurbit. Iowa State College Journal of 
Science 12:253-255. 

FLANNERY, KENT (editor). 1986. Guila 
Naquitz. Archaic Foraging and Early 
Agriculture in Oaxaca, Mexico. Aca- 
demic Press, Orlando. 

GILMORE, MELVIN R. 1931. Vegetal re- 
mains of the Ozark bluff-dweller cul- 
ture. Papers of the Michigan Academy 
of Science, Arts, and Letters 14:83-102. 

HARPER, F. (editor). 1958. The Travels of 
William Bartram; Naturalist's Edition. 
Yale University Press, New Haven. 



Q 



1955. Vascular Plants of Illinois. Illinois 
>tate Museum Scientific Papers 6, 

Springfield. 

KAY, MARVIN, FRANCIS B. KING, and 
C. ROBINSON. 1980. Cucurbits from 
Phillips Spring: New evidence and 
interpretations. American Antiquity 
45:806-822. 

KING, FRANCIS B. 1985. Early cultivated 
cucurbits in eastern North America. 
Pp. 73-98 in Prehistoric Food Produc- 
tion in North America. Richard I. Ford 
(editor). Anthropological Papers 75, 
Museum of Anthropology, University 
of Michigan, Ann Arbor. 

MOHLENBROCK, ROBERT and DAVID 
LADD. 1977. Distribution of Illinois 



Southern 



Press 



NEE, MICHAEL. 1990. The domestication 
of Cucurbita (Cucurbitaceae). in New 
Perspectives on the Origin and Evolu- 
tion of New World Domesticated Plants. 
Peter K. Bretting (editor). Economic 
Botany 44(3 SUPPLEMENT):56-68. 

NEWSOM, LEE A., S. DAVID WEBB, and 
JAMES S. DUNBAR. 1993. History and 
geographic distribution of Cucurbita 
pepo gourds in Florida. Journal of Eth- 
nobiology 13:75-97. 

OLIVER, L., S. HARRISON, and M. 
MCCLELLAND. 1983. Germination of 
Texas gourd (Cucurbita texana) and its 
control in soybeans (Glycine max). Weed 
Science 31:700-706. 

PITCAITHLEY, DOUGLAS. 1987 Let the 
River Be: A History of the Ozark's Buf- 
falo River. Southwest Cultural Re- 
sources Center, National Park Service, 
Santa Fe. 



54 



COWAN & SMITH 



Vol. 13, No. 1 



SIMPSON, GEORGE G., ANNE ROE, and 
ROGER C. LEWONTIN. 1960. Quan- 
titative Zoology. Harcourt, Brace, and 
Co., New York. 

SMALL, JOHN K. 1918. Narrative of a 
cruise to Lake Okeechobee. American 
Museum Journal 18:685-700. 

1922. Wild pumpkins. Jour- 
nal of the New York Botanical Garden 
23:19-23. 

1930. The Okeechobee gourd. 

Journal of the New York Botanical Gar- 
den 31:10-14. 

SMITH, BRUCE D. 1987. The independent 
domestication of indigenous seed bear- 
ing plants in eastern North America. 
Pp. 3-47 in Emergent Horticultural 
Economies of the Eastern Woodlands. 
William F. Keegan (editor). Occasional 
Paper No. 7, Center for Archaeological 
Investigations, Southern Illinois Uni- 
versity, Carbondale. 

, C. WESLEY COWAN and 

MICHAEL P. HOFFMAN. 1992. Is it 
an indigene or foreigner? Pp. 67-100 in 
Bruce D. Smith: Rivers of Change: Es- 
says on Early Agriculture in Eastern 



North America. Smithsonian Institu- 
tion Press, Washington, D.C. 
SMITH, DAVID V. 1992. The amino acid 
content of seeds of pre-maize crop 



Manuscript 



Museum 



EDWIN 



Washingt 



notated Checklist of the Vascular Plants 
of Arkansas. University of Arkansas 

Press, Fayetteville. 

. 1988. An Atlas and Anno- 

tated Checklist of the Vascular Plants 
of Arkansas. Second Edition. Univer- 
sity of Arkansas Press, Fayetteville. 

SPAULDING, P. 1909. A Biographical His- 
tory of Botany at St. Louis, Missouri. 
Popular Science Monthly 74. 

STEYERMARK, JULIAN. 1963. Flora of Mis- 
souri. Iowa State University Press, Ames. 

WALTERS, TERRENCE W. and DEENA S. 
DECKER- WALTERS . 1993. Systematics 
of the endangered Okeechobee gourd 
(Cucurbita okeechobeensis: Cucurbitaceae). 



Systematic Botany 18:175-187. 



WILSON 



449- 



^™ 



/. Ethnobiol. 13(l):55-72 



Summer 1993 



ISOZYMIC CHARACTERIZATION OF 
WILD POPULATIONS OF Cucurbita pepo 



DEENA S. DECKER-WALTERS 

Fairchild Tropical Garden 

11935 Old Cutler Road 

Miami, FL 33156 



TERRENCE W. WALTERS 

Fairchild Tropical Garden 
11935 Old Cutler Road 



M ia m 



)f Biological 



Florida International University 



Miami 



C. WESLEY COWAN 
it i Museum of Natural 

1720 Gilbert Ave. 
Cincinnati OH 45202 



BRUCE D. SMITH 
National Museum of Natural History 

Smithsonian Institution 
Washington, D.C. 20560 



ABSTRACT. -Isozyme data were collected from 20 wild populations of Cucurbita 
pepo ssp. ovifcra var. ozarkana recently discovered in Arkansas, Illinois, Kentucky, 
Louisiana, Missouri, and Oklahoma. Comparison of these data to similar data 
generated for populations in Texas (ssp. ovifera var. texatia) and Tamaulipas, Mexico 



frate 



ifera var. ovifi 



genetic lineages ^ssp. uuijtru \<xi. uuijcru an« jj^. y^r~> 

profile, including the characteristic allele Idh-2m, for the non-Texas U.S. popula- 
tions. About half these populations exhibited signs of limited and, in many cases, 
recent introgression from cultivars. Genetic similarity and the patterns ofvana- 

. . .. . 3 ;/„~„ \^^A no fr» rnnrhiriP that 



ifera var. ozarkana and var. ovif 



fraterna 



Mexico also exhibited 



iiuih nortneastern lviexico aisu caiuuhcu a un^uv, -^v; — r — 

to both major cultivar lineages. None of the wild populations exhibited a P art,cu " 
larly close relationship to Mexican landraces of ssp. pepo, leading us to hypoth- 
esize that progenitor populations of these landraces probably occurred farther 
south in Mexico and may be extinct. In conclusion, the isozyme data revealed that 
within the range of wild C. pepo, genetic divergence took place long before domes- 
tication and over an extensive period of time in at least four disjunct and ecologi- 
cally distinct regions-the central Mississippi valley/Ozark Plateau, Texas, north- 



eastern Mexico, and central or southern Mexico. 



56 DECKER-WALTERS, WALTERS, COWAN & SMITH Vol. 13, No. 1 



RESUMEN— Se colectaron datos de isozimas de 20 poblaciones silvestres de C. 
pepo ssp. ovifera var. ozarkana descubiertas recientemente en Arkansas, Illinois, 
Kentucky, Louisiana, Missouri y Oklahoma. La comparacion de estos datos con 



fraterna) 



'/< 



mente (ssp. pepo), y cultivares que representan los dos linajes geneticos prin- 
cipales (ssp. ovifera var. ovifera y ssp. pepo) revelo un perfil isozimatico distintivo, 
incluyendo el alelo caracteristico ldh-2m, para las poblaciones de los Estados 
Unidos de Norteamerica fuera de Texas. Alrededor de la mitad de estas pobla- 
ciones mostraron senas de una introgresion de cultivares, limitada y en muchos 
casos reciente. La similitud genetica y los patrones de variacion entre la ssp. 
ovifera var. ozarkana y la var. ovifera nos conducen a concluir que las poblaciones 
antiguas de la primera dieron origen a la segunda. Las poblaciones de la ssp. 
fraterna del noreste de Mexico tambien mostraron un perfil isozimatico peculiar 
con afinidades a los dos linajes principales de cultivares. Ninguna de las pobla- 
ciones silvestres mostro una relation particularmente cercana a las variedades 
mexicanas tradicionalmente cultivadas de la ssp. pepo, llevandonos a la hipotesis 
que las poblaciones progenitoras de estas variedades probablemente se daban 
mas al sur en Mexico y posiblemente se hayan extinguido. En conclusion, los 
datos de isozimas revelan que dentro del area de distribution de la C. pepo 
silvestre, la divergencia genetica tuvo lugar mucho antes de la domestication y a 
lo largo de un extenso periodo de tiempo en por lo menos cuatro regiones 
geograficamente separadas y ecologicamente diferentes: la region central del 
Valle del Mississippi junto con la altiplanicie Ozark, Texas, el noreste de Mexico, 
y el centro o sur de Mexico. 

RESUME. -Des donnees d'isozymes ont ete rassemblees, provenant de 20 pop- 
ulations sauvages de C. pepo ssp. ovifera var. ozarkana recemment decouvertes 
dans 1' Arkansas, l'lllinois, le Kentucky, la Louisianne, le Missouri, et TOklahoma. 
Des comparaisons faites entre ces donnees et celles obtenues sur des populations 
du Texas (ssp. ovifera var. texana) et de Tamaulipas, Mexico (ssp. fraterna), des races 
mexicaines (ssp. pepo), et des "cultivars" representant les deux lignees genetiques 
principales (ssp. ovifera var. ovifera et ssp. pepo) ont revele un profile d'isozymes 
distinct, comprenant le gene (allele) characteristique ldh-2m, pour la population 
non-Texanne des USA. A peu pres la moitie de ces groupes exhibe des signes 
d'mtrogression limitee, et, dans plusieurs cas, recente, de "cultivars." La resem- 



'/< 



ifi 



fraterna 



Mexique exhibent aussi un profile d'isozyme unique, exhibant des affinitees avec 
les deux lignees principales. Aucune des populations sauvages exhibent une 
parente particulierement proche des races de ssp. pepo du Mexique, ce qui nous 
amene a supposer que les populations ancestrales de ces races se trouvaient 
probablement plus au Sud du Mexique, et sont peut-etre eteintes. En conclusion, 
les donnees d'isozymes ont revele qu'au sein de Letendue de C. pepo sauvage, une 
divergence genetique a pris place longtemps avant la domestication, s'etendant 
sur une longue periode, dans 4 regions non-adjacentes et ecologiquement dis- 
tinctes-la vallee du Mississippi central/region du Plateau Ozark, le Texas, le 
Nord-Est du Mexique, et le centre ou le Sud du Mexique. 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 57 



INTRODUCTION 



a 



As one of the earliest New World domesticates, squash (Cucurbita pepo L.) has 
great impact on theories of horticultural origins in North America. A greater un- 
derstanding of the evolutionary history of this species has led to the suggestion 
that squash domestication took place independently twice to produce the two 
major lineages of cultivars, C. pepo ssp. ovifera (L.) Decker var. ovifera (L.) Decker 
nd C. pepo ssp. pepo (Decker 1988). Subspecies pepo, which comprise^ pumpkins, 
marrows, Mexican landraces, and a few ornamental gourds, apparently has its 
origins in Mexico, whereas ssp. ovifera var. ovifera, which includes scallop and 
crookneck squashes and most ornamental gourds, appears native to the eastern 
United States (Decker 1988; Decker-Walters 1990). In the past, examination of wild 
populations that might have given rise to the cultivar lineages was limited to 
several populations of C. pepo ssp. ovifera var. texana (Scheele) Decker in Texas and 
a few morphologically similar populations in Alabama, Arkansas, Illinois, and 
Missouri (Decker and Wilson 1987). Isozyme analyses indicated that these popula- 
tions were remnants of ancestral wild C. pepo and were not merely cultivar escapes 
(Decker and Wilson 1987). Consequently, var. texana became the likely candidate 
for the progenitor of the var. ovifera cultivars. Although isozyme data were lacking 
at the time, Decker-Walters (1990) hypothesized that C. pepo ssp. fratema (Bailey) 
Andres, which occurs in northeastern Mexico, was the wild progenitor of the ssp. 
pepo lineage of cultivars. No other wild taxa of C. pepo are currently recognized. 

Recently, a few populations of ssp. fratema became available for isozyme analy- 
sis (cf. Wilson 1989). Additionally a much greater number of wild populations oc- 
curring north of Texas were discovered by various collectors (Smith et al. 1992). 
Because the solid ivory fruits of these northern populations differed from the 
striped fruits of var. texana and ssp. fratema, we decided to look for other genetic 
evidence distinguishing these populations. Therefore, we collected isozyme data 
on these and other populations to clarify further the evolutionary relationships 
within C. pepo, particularly those between wild populations and the two domesti- 
cated lineages. 

MATERIALS AND METHODS 

For purposes of this study, all wild populations occurring north of Texas were 
considered unclassified, whereas only those populations in Texas were treated as 
belonging to Cucurbita pepo ssp. ovifera var. texana. _ 

We surveyed isozymes 
north and east of Texas, six \ _ r . 
fratema, four cultivars of ssp. ovifera var. ovifi 



'f 



landraces of ssp. pepo (Tables 1 and 2; Fig. 1). In some analyses, we considered 
previously collected isozyme data on six non-Texas U.S. populations (Decker and 
Wilson 1987), 12 additional populations of ss P . ovifera var. texana (Decker and Wilson 

-id approximately 20U 

ars) representing ssp. 



culti 

ovifera var. ovif 



fratema (Wils 



Walters et al . 1990 



ovifera var. ovifera and ssp. pepo (Decker iv»d, i too, i^*c. . .««.*- * -• - - ? 
and Weeden 1984; Wilson 1989). Existing data on the closest extant wild relative -or 
•- -, „ , °. /n„;i^,\ \4orru-Lr and Rates Decker 1986; 



C. r . 

Wiis< 

pepo . 



58 



WALTERS, WALTERS, COWAN 



Vol. 13, No. 1 



< 



^ 

S 



en 

c/) 

bJD 






v 



<o 



u 



G 

C/) 

c 

O 



12 

o 



T5 



* 




Cn 

— 

CD 

E 
2 

a> 
c 

cd 
cd 



a» 



x 



O 

2 



3 



cd 
- 



cn 



O 

2 



12 

3, 



CD 

— 



X 



5 T3 

2 c 



£ 



01 

> 



C£ 



cd 

CD 

u 



>> 



C 

u 



CD 

13 



^ 



c 
U 



E 



<n 

3 



I 



CQ 
< 



vO 



rv 



GO 

fN 

(N 

00 
CM 

CO 

04 



rO 
rN 

m 
ro 

(N 
(N 

IN 



ro 



(N 
in 



(N 



CO 

CO 
CN 

IN. 
CM 

in 

ro 

CO 



o ^o 

CM 



CQ CO 
< < 



CO 

ON 

ro 
ro 



IN. 

ro 



in 
tN 

IN. 
(N 

IN. 
IN. 



<n 

ON 



ON 

CO 



(N 

ro 
lo 

ON 



CO 



ON 

IN 

CQ 
< 



i 

ro 

CO 



ON 

(N 
i 

CO 

CO 

rN 



CO 
IN. 

ro 

tN 

ro 

i 

ro 

IN. 

ro 

ro 
i 

<* 

ro 

(N 

NO 
ro 



"J* 

i 

CO 

ON 

ro 

ON 

ro 

ON. 

ro 
r\i 

ON 

ro 



i 

LO 

ro 



CO 

rN 



rN 



LO 



rN 
rN no 



ON 



00 O N 



CQ 
< 



CQ CQ 
< < 



CQ 
< 



CQ CQ 
< < 



ON 

ro 

i 

IN. 
CO 

ro 
CQ 
< 



CO 

on 

£! ^ o 

°0 CO ON 

o m \d 

CO CO CO 

CQ CQ CQ 
< < < 



ON 



I 
LO 



IN. 



CO 



rN 

LO 



ro 



LO 

l 



CQ 
< 



CQ 
< 



CQ CQ 
< < 



ON 
LO ON 

CQ CQ 
< < 



LO 

ro 

rN 

ro 



LO 

ro 
rN 

ro 
lo 



ro 



D- CL, 



rN 

LO 

CO 
(N 

• X 

■1 W 

a, P 



n ro rf in 

X X X X 

WW WW 

b* t-* H H 



o co on on co rN 



ro rN O on o tN 



Tf rN co lo 



co 



fs.. fv. fv. f^« 

rf N X X 



f^. fV 



LO rN r^. 



+ 
rv. <%.. f^. fv.. (r^ f\j 



r-. f^. r< 



r^- 



r^» r^ 



ocooNascooaNcoroavooaNro^ONrNcoooN 



00 CO CO CO CO 



03 it! 

O » 



C LT) 



> ^ 

03 t: 

Qu.t: 

en w 



03 



o <u 



3 

CQ 



x 



C 



a. 

en 





a: 2 



c£ 



c 
g 



en 

cd 



a e 




03 



C 

03 



0> 

C 

o 

03 
U 



03 

C 

o 

03 



CJ 



.2 

03 T3 

^^ 
CQ > 



5 







3 

03 
03 

C3 

u 



G 



en 

8 .2 
< S 



< 

03 

en 



03 03 

3 O 
(J U 



o o c 







C OS C 

a» a> a> 

cq en DC: 



c 

CQ 



03 

LO 



c 

a; 



03 

N 



C 






03 

cn 
03 

C 



.Si 
"55 

en 

G 



03 



en 




C 

CD 



u a 



at 
C 

03 

H 



OJD oc fc 

^ ^ o 



CD 

O 

>^v CL» 



Su 



03 

< 



en 

03 

B 

03 

6 

n3 



en 

03 

CX en 

03 N 
C 

03 





u 



H H U 



03 
CU 

c 

03 

en 



I "8 

C3 -z: 



a. o 



en 

*03 

N 

C 
O 

U 



a> 



CD 

>^ 

03 

W 



^ 5 ^ ^ K ^ ^ 




^^^Q§§Q92^^^WWXXXXXX 




gggcoosorj^r-^xoN 



cd m ^o rN m \o in. o 



[^ C^ Q 

LO ^t N X 

i\ in r-x r-. 



00 



00 CO 



< 



-J b < 

== >^ w 

r-. rv. r^. 



o 




•^ 



o 



< 

E 



x 



- 



OJ 




c 



I 



a 





C/5 

cv 

1 

c 






»- 

1 

1 

i 



C 



5 

X 

- 

Z 



c 
E 

c 
c 

"3 

z 



c 

5- 

c 

(B 

-J 

a 

£ (0 



X 

3" 



5 



w 
- 

^ 



- 



X 5 



C 



*3 
/ 

i 

C / 

■^ - 

IS 



u 

I 

c 

1- 



C 

§1 



= 



sb. 



Summer 1993 



JOURNAL OF ETHNOBIOLOCY 



S9 



TABLE 2.— Accession and sampling information on eight cultivars and tour Mex- 
ican landraces of Cucurbita pepo. 



Taxon 



Accession 1 



Cultivar/ 
Landrace 



Sou ret* 2 



No. 

individual^ 
sampled 



ssp. ouifera 
var. ovifera FMA 467 

OEN 538 
OFS 774 
OPS 763 



ssp. pepo 



'Mandan' 
'Nest Egg' 
'Flat Striped' 
'Striped Pear' 

'Orange Ball' 



OBO 775 

OWO 765 'Orange Warted' 



Nichols Garden Nurserv 
Stokes 

J.L. Hudson 

Stokes 
J.L. Hudson 



PCO 759 
PSU 772 
XGU 777 
XHU543 
XQU 778 
XYU 468 



'Connecticut Field' J.L. Hudson 



'Small Sugar' 
Mexican landrace 
Mexican landrace 
Mexican landrace 
Mexican landrace 



Stokes 
P.I. 512190 
Stokes 
P.I. 512192 



11 

B 

8 

9 

10 
8 

8 

8 
8 

S 

5 



'Accession numbers are the collection numbers of D. Decker-Walters and T. Walters. 

2 All seed was obtained from seed companies or the U.S.D.A., except tor FMA 467 which was contrib- 
uted by G. Drowns, and XYU 468, which is F, seed of a self of P.I. 438699 created by T. Andres 

The following northern populations showed morphological signs (e.g., non- 
bitterness, large seeds, thick peduncles) of hybridization with cultivars: ?AR802, 

->AT-»o-ir» -.T^x/r,-^ -.t ^^* . -»»*^r>-i/' /o_:i.u ~* -I IQQ-)- ffHvan xnA Smith this 



?M0816 



Morpholo 



VVilso 



We examined 



vve examined isozymes in rive to i:> muiviuuuis pci »w«~.. v / 

Enzymes were extracted from 3- to 5-day-old cotyledons as described in Decker- 

UJ i. * - .., , i . i CC •^Xama fr% rocnKp fhp following 



Walters et al. (19901 We 



me systems 



6.7) system, isocitrate 



malate dehydrogenase (MDH) 



utnyurogenase (ilji-ij, maiate aenyarugenct^c vivil^a *» K ..*^ r .-~ - 

and shikimate dehydrogenase (SKD); and on the Poulik system (system #1 of 

L^-i . , , «;£*»* ° . . • „ t ^.m/AAT\ alvrprate dehvdroee- 



aminotr 



*MiApciLllLK t?t dl. LVOD), dbpdlldlt: dlini iuuciiwi^»"^ v " o J ' ~ 

nase (G2D), glucose-6-phosphate isomerase (GPI), and leucine aminopept.dase 
(LAP). Recipes and additional details on the electrophoretic methodology are de- 
scribed elsewhere (Decker- Walters et al. 1990; Kirkpatrick et al. 198:?). 

Genetic interpretation was based on the nature of the banding patterns and 



rformed 



mber of loci detected for some enzymes (e.g., POM) 



P o 



r 



the genus Cucurbita. However, 



piugeny segregation analysis connrmeu 

ically as diploids (Kirkpatrick et al. 1985). 

We resolved allelic variation at 18 e 



behav 



Lap-l Mdh-1, Mdh-2, Mdh 



60 



WALTERS, WALTERS, COWAN 



Vol. 13, No. 1 




sam 



previous study (solid symbols; Decker and Wilson 1987). Triangles represent Cu- 
curbita pcpo spp. fratema, squares represent ssp. ovifera var. texana, and circles rep- 
resent unclassified wild populations. FIGS. 2-4.— Solid symbols indicate pres- 
ence of a particular allele, with the larger (if fhp^P indiraHne 100% frequency 



represent ssp. ovifera var. texana, and circles represent 
tions. FIG. 2-ldh-3o. FIG. 3.-Aat-le: FIG. 4.-hih-2m. 



fratema 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 



hi 



Pgm-6, Skd-1. Pgm-6 is equivalent to Pgm-2 of previous reports (e.g., Decker 1985; 
Wilson 1989). We designated alleles in accordance with Kirkpatrick et al (1985) 
and Decker- Walters et al. (1990) with the following exceptions: Idh-lg of previous 



renamed 



comigrat 



Walters 



1990) as a null or inactive variant (his hih-2z). Because either interpretation is 



Walters 



ave 



at the same time recognizing that the null interpretation may be correct. Our null 
variants are denoted with a superscripted "n" (e.g., Lap-lf"). 

The ability to detect first-generation hybridization is one of the advantages of 
isozyme data. Heterozygous conditions are clearly visible and when an individual is 
heterozygous for alleles typically not found in that taxon, it is often possible t< i 
determine the taxon that contributed the "foreign" pollen. The opportunity to doc- 
ument infraspecific hybridization is greater in Cucurbita pepo than in most species 
because there are at least two genetically distinct but interfertile lineages: ssp pepo 
and ssp. ovifera (Decker 1988). Consequently, we looked for signs of first-generation 
hybridization as well as longer histories of introgression in the isozyme data on the 
wild populations. First-generation hybridization could be postulated for an indi- 
vidual exhibiting a larger than normal percentage of heterozygous loci at which one 
of the alleles per locus rarely or never occurred in other members of the same 
population and/or taxon. Introgression beyond the first generation was suspected 
when one of these apparently foreign alleles was found in the homozygous state. 

Other data analyses also focused on describing wild populations north of 
Texas and on comparing these populations to infraspecific taxa of C. pepo and to 
C. argyrosperma ssp. sororia. We mapped alleles in wild populations of C. pepo to 
look for geographical trends. Genetic affinities within C. pepo were illustrated via 
the plotted results of principal component analysis (PCA), which was performed 
using the Statistical Analysis System (SAS Institute Inc. 1985). Following Nei 
(1972), as in previous studies, we calculated genetic identity values within and 
among taxa using the data presented herein, except for data on C. argyrosperma 
ssp. sororia, which came from Decker (1986). 

Drawing on all available sources of isozyme data (see above), we attempted to 
characterize each taxon, including the unclassified populations occurring north of 



We 



Mdh-2. Mdh 



Pgm-6. Alleles found rarely or as apparent introgressants in a taxon were not 
included in that taxon's isozyme profile. Introgressants were presumed based on 
heterozygous conditions such as those described in general terms, above, and 
more specifically, below. 



RESULTS 



The f req 



Mdh-lc, andPw?-?fl. M 



widespread across accessions and therefore uninformative (e.g., G2d-ln, Idh-le, 
Mh-2g, Lap-lf, Mdh-2i, Pgm-6v). 



62 



DECKER-WALTERS, WALTERS, COWAN & SMITH 



Vol. 13, No. 1 



TABLE 3. — Frequencies (expressed as percentages) of 40 alleles at 15 enzyme loci for 
28 wild populations, eight cultivars, and four Mexican landraces of Cucurbita pepo. 



Locus 


Aat-1 


T 

Aat-3 


Aat-4 


G2d-1 


G2d-2 


Gpi-3 


Idh-1 


Idh-2 


Allele 


b d e 


s i 


r 


11 


V 


I 


k n 


V 


X" 


u 


b e 


g »< 


Wild populations: 
























7AR801 


25 75 


100 


100 











100 


100 





100 


100 


100 


7AR802 


45 55 


100 


86 


14 





5 


95 


100 





100 


100 


55 45 


7AR803 


5 95 


100 


100 











100 


100 





100 


100 


75 25 


?AR804 


100 
89 11 


100 


100 












100 


100 





100 


100 


94 6 


7AR818 


100 


100 








100 


100 





100 


100 


89 11 


7AR819 


39 61 


100 


100 








11 


89 


100 





100 


100 


33 67 


7AR820 


50 50 


100 


63 


12 


25 





100 


100 





100 


100 


81 19 


7IL812 


100 


100 




1 00 


100 








100 


100 





100 


100 


100 


7KY813 


100 


100 








39 


61 


100 





100 


100 


100 


7LA811 


100 


100 


88 


12 








100 


100 





100 


100 


100 o 


7LA814 


100 


100 


100 








27 


73 


100 





100 


23 77 


100 


7MO808 


44 56 


100 
100 


89 


11 









100 


100 





100 


100 


67 33 


7MO809 


50 50 


100 








100 


100 





100 


100 


50 50 


7MO810 


40 60 


100 


100 











100 


100 





100 


100 


80 20 


7M0815 


22 78 


22 78 


100 











100 


100 





100 


100 


56 44 


7M0816 


15 58 27 


8 92 


100 











100 


100 





100 


100 
4 96 


65 35 


7M0822 


100 


4 96 


100 











100 


100 





100 


18 82 


7OK805 


100 


100 


100 











100 


100 





100 


100 


100 o 


7OK806 


63 37 


100 


100 











100 


100 





100 


100 


96 4 


7OK807 


44 56 


100 


100 
100 












100 


100 





100 


100 


88 12 


FRA750 


100 


100 





15 85 


100 





35 65 


20 80 


1 00 o 


FRA751 


100 


100 


100 











100 


100 





100 


100 


100 o 


TEX747 


100 


100 


100 








100 


100 





100 


100 


100 o 


TEX779 


1 00 
1 00 


100 


88 


12 
100 









100 


100 





100 


100 


100 o 


TEX780 


100 


o: 





100 


100 





100 


100 


100 o 


TEX781 


100 


100 100 








100 


100 





100 


100 


100 o 


TEX782 


100 


100 


100 











100 


100 





100 


100 


100 o 


TEX783 


100 


100 


56 


44 








100 


100 





100 


100 


_100__J_ 


var. ovifera: 












FMA467 


50 50 


100 


50 


50 





36 


64 


100 





100 


28 72 


100 o 


OEN538 


12 88 


69 31 


100 











100 


94 


6 


25 75 


6 94 100 


OFS774 


100 


100 


100 











100 


100 





6 94 


100 


100 o 


OPS763 


100 


100 


100 











100 


100 





100 


too 


_100_J_ 


ssp. pepo: 
























OB0775 


22 78 


33 67 


100 








6 


94 


100 





17 83 


17 83 


100 o 


OW0765 


55 45 


100 


100 











100 


100 





100 


5 95 


100 o 


PC0759 


69 31 


100 


100 








6 


94 


94 


6 


94 6 


94 6 


100 o 


PSU772 


75 13 12 
1 00 


100 


94 


6 









100 


100 





100 


100 __o 

~~94 6 


100 o 


XGU777 


100 


100 








100 


100 





1 00 


100 o 


XHU543 


100 


97 3 


100 











100 


100 





100 


83 17 


1 00 o 


XQU778 


100 

V 


100 


100 











100 


100 





100 


25 75 


100 o 


XYU468 


100 


100 


100 











100 


100 





100 


1 00 o 


100 o 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



63 



Idh-3 



m o 



Lap-1 



e f f" S 



Mdh-2 



T 



/ 



Mdh-3 



T 



in 



q 



u 



Pgm-5 Pgm-6 



Skd-1 



n 



o 



v 



g h I k in p 



100 

100 

100 

100 



100 

100 

85 15 

100 



85 15 

18 82 

25 75 

37 63 



100 

14 86 

5 95 

100 



100 
100 
100 
100 



100 

100 

100 

100 






100 

44 56 

85 15 

100 



100 

100 

100 

100 



100 

100 

100 

100 



100 
100 

100 
12 88 

100 

100 

100 

100 



100 

89 11 

100 

100 



100 

56 44 

100 

100 



100 

100 

83 17 

8 92 



96 4 

100 

100 

100 



28 72 

33 67 

100 

100 



11 89 

61 39 

25 75 

20 80 



56 44 

100 

100 

44 56 



56 44 

100 

100 

11 89 



50 50 
100 

17 83 
4 96 



100 
100 
4 96 
100 



40 60 

61 39 

100 

100 



100 

100 

100 

100 



100 

100 o 

100 o 

100 o 



100 

100 

100 

100 



100 

19 62 19 

81 19 

88 12 



17 83 

100 

19 6 75 

50 50 



100 

100 

100 

100 



100 
100 
100 
100 



100 

44 56 

100 

100 



100 50 50 



100 
100 
100 



73 27 
100 
100 



100 

100 

100 

19 81 



100 
100 
100 
100 



89 11 



100 

100 

72 28 

77 23 



82 18 

100 

100 

100 

100 

100 

100 

100 



100 

100 

100 

100 



100 

38 6 56 

37 44 19 

100 







25 5 70 
25 75 
100 



100 

100 

100 

100 



100 
100 
100 
100 



100 
100 
100 
100 



100 
100 
100 
100 



100 
100 
100 
100 



100 
100 
100 
100 



100 
100 
100 
100 



9 91 
100 
100 
100 



100 

10 90 

100 

100 



100 
100 
100 
100 



100 

5 95 

100 

37 63 



100 
100 
100 
100 



100 

100 

100 

100 



50 50 
100 
100 
100 



100 

100 

17 83 

100 



100 
100 
100 
100 



100 

72 28 

37 63 

100 



100 

6 94 

100 

22 78 



75 25 

30 70 

100 

4 96 



4 96 

100 

100 

6 94 



90 10 
6 94 
100 
100 



100 
100 
100 

100 



100 

14 86 

6 94 

100 



100 

100 

100 

100 



100 

100 

100 

100 



94 6 

100 

81 19 

75 25 



100 

83 17 

100 

100 



6 94 

50 50 

19 81 

56 44 



100 

75 75 

69 13 12 6 

38 6 37 19 



75 25 

22 11 67 

94 6 

100 



64 DECKER-WALTERS, WALTERS, COWAN & SMITH Vol. 13, No. 1 



Some alleles indicated a genetic link between populations in Texas and those 



_, „ . Wilson „ _ 

allele that characterizes var. texana (Table 3; Fig. 2). It has never been found in any 
cultivar or landrace of var. ovifera or ssp. pepo (Table 3; Decker 1986; Wilson 1989). 
However, this allele was detected in both populations from Louisiana (Table 3) and 
in a population from southwestern Arkansas (Fig. 2; Decker and Wilson 1987). The 
allele Aat-le- exhibited a similar pattern; it characterizes populations of var. texana, is 
rare in domesticated material (Decker 1986), and was found in several of the popula- 
tions north of Texas (Fig. 3). In fact, three non-Texas populations (7AR801, 7M0822, 
?OK805) were fixed for Aat-le- (Table 3; Fig. 3). Another allele linking Texas popu- 
lations to those farther north is AatAu (Table 3). We found this allele in all but one 
Texas population, in five populations to the north (7AR802, 7AR820, 7IL812, 
7LA811, 7MO808), and in two cultivars (FMA467, PSU772). However, since only 
cultivars in this study have been tested for variation at Aat-4, the distribution of 

- 11 ^1 _//..//• J _ . • i . . * . 



allele " u 



determined 



Within C. pepo, a few alleles, including Aat-4ir, Idh-2m, and Lap-lf', were unique 
to populations north of Texas (Table 3). Of these, only Idh-2m was widespread, 
characterizing populations in Arkansas, Missouri, and northwestern Louisiana 
(Fig. 4). This allele, which was interpreted by Wilson (1989) as a null or inactive 
variant, appears to occur as a dominant allele in C. argyrosperma as well (Decker 
1986; Decker-Walters et al. 1990; Wilson 1989). If introgression from C. argyrosperma 
accounted for the widespread presence of ldh-2m in populations of C. pepo, then we 



Walters 



Lap 



(Table 3). Remaining possible explanations are that ldh-2m was either present in the 
shared ancestor of these two species or had separate origins in these species or their 



The latter interpretation 



many 
allele has not been found in cultivars of C. pepo 



Table 3 quickly reveals that the unclassified northern populations exhibited 



ifera and ssp. hate 



Wilson 



et al. 1992) already indicated that ssp. pepo has followed a distinct evolutionary 
path, alleles that characterized ssp. pepo and occurred in few wild populations, 
usually as one or two heterozygous individuals, were assessed as cultivar intro- 



Mdh- 



M, i gm-bs, and bkd-lg ( Table 3). Using these alleles as markers, we de 
generation hybridization resulting from cultivar pollen on wild stigmas 
7LA814, 7MOS22, and 7OK807. For example, the genetic configuration 
vidual in 7M0822 included Aat-3gj, Idh-lbe; Lap-lfg, and Skdj-lim, v 
remaining individuals, including another from the same fruit, were h. 



comm 



m 



form 



Wilson 



of 



surve 



for 



?M0815, and ?M0816 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 65 



TABLE 4.— Isozyme profiles for taxa of Cucurbita pepo and C. argyrosperma spp. 
sororia. Data from other sources (e.g., Decker 1986; Wilson 1989) were considered 
also. Acorn cultivars were not included in C. pepo spp. ovifera var. ovifera, whereas 
ssp. pepo was represented by Mexican landraces only. 

Aat-1 Gpi-3 Idh-1 Idh-2 Idh-3 Mdh-2 Mdh-3 Pgm-6 

C. pepo 

var. ovifera d u e~ g m eimqu v 

northern populations de- u egmm ei mq v 

var. texana e- u e - g mo i q 

ssp. fraterna d oube-g m i q sv 

ssp. pepo b o be- g m i usv 

C. argyrosperma 
ssp. sororia e~ o e - g m m i q 



TABLE 5.— Nei's (1972) genetic identity values (percentages) calculated for 
C. pepo ssp. ovifera var. ovifera, ssp. ovifera var. texana, ssp. fraterna, spp. pepc 
and unclassified northern populations. 



Taxon 



O ? T F P 



var. ovifera (O) 89 

northern populations (?) 89 93 



var. texana (T) 
ssp. fraterna (F) 
ssp. pepo (P) 



79 86 92 

84 85 76 95 

68 64 56 71 87 



?M0816, for example, one individual was homozygous for Up-lee and Mdh-3uu 
and heterozygous for Aat-lbd. In each of these cases the putative introgressant 
alleles occurred within the population at a frequency less than 30% (Table 3) ex- 
cept for ?KY813, which had high frequencies of Mdh-3u (44%) and Pgm-6s (50%), 
suggesting that this fruit may have come from a feral domesticated plant . 

One allele that was surprisingly missing from northern populations was Gpi- 
3o. This allele, which is dominant in ssp. pepo cultivars (Table 4; Decker 1986, 1988), 
would be expected to occur in wild populations if cultivar introgression were 



common 



mor 



One reason why foreign alleles were easily recognizable is because most of the 
populations north of Texas possessed a unique and largely coherent genetic pro- 
file (Table 4). This profile is most like that of the ssp. ovifera var. ovifera lineage of 
cultivars, with some similarity to var. texana and ssp. fraterna as well (Table 4). 

These profiles along with the genetic identity values (Table 5) suggest rela- 
tionships among the other taxa. The taxon most genetically similar to the ssp. pepo 
lineage is ssp. fraterna (Table 5). However, there are significant differences be- 
tween these taxa at Aat-1 and Mdh-3 (Table 4), and ssp. fraterna exhibits larger 
genetic identity values with ssp. ovifera than it does with ssp. pepo (Table 5). 



66 DECKER- WALTERS, WALTERS, COWAN & SMITH Vol. 13, No. 1 



3 



2- 








CVJ 1 




o 



CD 



o 




E 
o 

O 

15 



o- 




o 




0_ 



O 



-H w o 



8 e° 



.A A 








A 



O unclassified 




-2 



o 




var. ovifera 
A ssp. f rate ma 

var. texana 
c Mexican landraces 




A 



ssp. pepo 



-3 



t . r 



-1 1 2 3 

Principal Component 1 

FIG. 5. -Plot of the first two principal components in the analysis of 20 unclas- 



\fera var. ovifi 



two 



populations of ssp. fraterna, six populations of ssp. ovifera var. texana, and four 
Mexican landraces and four cultivars of ssp. pepo. 

Taxon relationships are also revealed in the plotted results of the principal 
component analyses (Figs. 5 and 6). In the first analysis (Fig. 5), which included all 

accessions, thp firsf rr»mnr»no«4- /T>/^t\ . • r .r, n , <■ .i .- t -l ..^^^.finn. 



component 



served to distinguish cultivars and Mexican landraces of ssp. pepo. Among mem- 
bers of ssp. pepo only 'Orange Ball' had a relatively low value for PCI (0.963), 

olacine it near 'Npst Voo' mn — n c^m i r ■• i. L = c rcn fra- 



placing it near 'Nest Egg' (PCI 
terna (PCI 



ssp. /i 



; /< 



from 



other accessions by having high values for PC2, which accounted for 13% of the 
total variation. The unclassified northern populations clustered more or less to- 
gether, except for the accession from Kentucky (PC2 = -1 839). 'Striped Pear' also 
had a low value for PC2 (-2.143), whereas 'Mandan' clustered with the northern 
populations and 'Flat Striped' occurred very close to one of the populations of 
ssp. fraterna. 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



67 



CO 

0) 



8. 

E 
o 

O 



03 




o 

c 



Q. 



4 



2- 



A 



A 




o 





0- 




o 
o 

° o° 
o o 





<o 




<b 



o 



o 



o 
o 



o 



-2- 



o 




o 



-4 




T 



-2 



-1 







T 
1 




unclassified 



var. ovifera 



A ssp. fraterna 



var. texana 



T 



2 



3 



Principal Component 1 



FIG. 6.— Plot of the first and third principal components in the analysis of 20 
unclassified populations of Cucurbita pepo, four cultivars of ssp. ovifera var. ovifera, 
two populations of ssp. fraterna, and six populations of ssp. ovifera var. texana. 



To further clarify the genetic affinities of domesticated ssp. ovifera, principal 
component analysis was performed after excluding domesticated ssp. pepo (Fig. 6). 
Along PCI, accounting for 30% of the total variation in this analysis, populations 



of 



ssp 



'/« 



sions. No distinct groupings were revealed by PC2 (not shown), which accounted 



fraterna 



Mand 



clustered with the bulk of northern populations. Lower PCI values were seen in 

'Flat Striped' (PCI = -0.850), 'Nest Egg' (-1.362), 'Striped Pear' (-1.663), and 

the fruit from Kentucky (-1.689). ?M0822 displayed the lowest value for PC3 (Fig. 6). 

In summary, the principal component analyses revealed the following genetic 



fraterna. Although, as a group, cultivars of ssp. ovift 



<* 



ifera did not appear to cluster randomly amon 



ther were they completely separated from these populations as a distinct grouping 
(Figs. 5 and 6). 



68 



WALTERS, WALTERS, COWAN 



Vol. 13, No. 1 





> 85% 




>75 




>65 



>50 



FIG. 7. 



Mexican landraces of ssp. pepo (M), ssp. ovifi 



among Cucurbita pepo ssp. fraterna 



sororia (S), and unclassified northern populations (Z). 



isozyme profile of C. argyrosperma 



Walters 



m 



L. pepo. In fact, the only uncertainties in this regard are Idh-2m, which was dis- 
cussed earlier, and alleles at Aat-1. Consequently, differences between C. 



argyro 



profiles 



prim 



genetic identity 



values (Mg. 7) suggest that C. argyrosperma ssp. sororw exhibits a closer relation- 
ship to one C. pepo taxon than it does to the other C. pepo taxa. In other words, 
divergence of each C. pepo taxon from the common ancestor of C. argyrosperma 
spp. sorona and C. pepo has occurred at very similar rates. Interestingly, genetic 
divergence between Mexican landraces of ssp. pepo and wild U.S. populations is 
greater than that between C. argyrosperma ssp. sororia and taxa of C. pepo (Fig. 7). 

DISCUSSION 

The status of northern populations .- As a group, wild populations of C. pepo north of 
Texas possessed a distinct though variable genetic profile, which usually included 
ldh-2m and the ssp. ovifera var. texana allele, Aat-le-. Louisiana and Arkansas popu 
lations close to Texas also exhibited an allele otherwise restricted to var. texana 

ufl- in Ihp n\7<=»ral1 icnvxrmn ~ £:i_ i ., — . .* _ 



isozyme 



evidence that 



,. «, /MHE K iuiut: ana tnese alleles in particular are evidence mm 

populations north of Texas developed their own genetic identity, probably over a 
lone period of re lative isolation 7 r 



long period of relative isolation. 



isozyme 



^ra, var. to™, and ssp. fraterna. However mo 
x fraterna is distinct from the remaining wild populations (Andres 
ample mature fruits of ssp. fraterna turn yellow-orange whereas th 
ana and the more northern populations do not. Also, ssp. fraterna 
atively slow to germinate (within 5-15 davs^ ,nH a* „!,„♦« a ~ f™,nH 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 69 



habitats as opposed to the riparian habitats characterizing var. texana and popula- 
tions to the north (i.e., Cowan and Smith, this volume). On the bases of fruit 
characters and ecogeography, we recognize wild populations north of Texas as 
more closely related to var. texana than to ssp. fraterna. Furthermore, ecogeo- 
graphic, fruit structure, and allozyme data indicate that these populations can be 
circumscribed as a distinct variety of ssp. ovifera, as described below. The circum- 
scription of var. texana is changed from Decker (1988) to include Texas populations 
only. 



Key to the Varieties of Cucurbita pepo ssp. ovifera: 

1. Fruit almost always bitter, solid ivory or green- and white-striped but never 
yellow or orange, rind smooth; seed germination within 1-7 days; possess- 
ing ldh-2m and/or Idh-3o; wild. 
2. Fruit typically solid ivory; germination within 1-4 days; 

possessing Idh-2m, may possess Idh-3o; Illinois, 

Missouri, Arkansas, Oklahoma, Louisiana. var. ozarkana 

2. Fruit typically green- and white-striped; germination 

within 3-7 days; possessing Idh-3o, lacking Idh-2m; 

Texas. var. texana 

1. Fruit non-bitter (except for some ornamental gourd cultivars), solid 
or striped, in a variety of colors including yellow and orange, rind 
smooth, ribbed, or warted; germination within 3-15 days depending 
on the cultivar; lacking Idh-2m and Idh-3o; under cultivation . var. ovifera 

Cucurbita pepo L. ssp. ovifera (L.) Decker var. ozarkana Decker-Walters, var. 
nov., OZARK GOURD.— TYPE: USA, Arkansas, Independence County, Nov 6, 



W. 



* 



seminum 



(Au- 



f 



at locus Idh-2 of isocitrate dehydrogenase and early seed germination, is distin- 



from 



Morphology. 



Mature fruits of C. pepo ssp. ovift 



ozarkana 



riform 



measunn 



mm 



mm wide. Deviations from 



evidence of cultivar introgression (Cowan and Smith, this volume). 

Isozymes. This variety is distinguished from other varieties of ssp. ovifera by 
possession of the isozyme allele ldh-2m. Populations of var. ozarkana are further 
characterized by the alleles Aat-ld and/or Ac, - 3j, -4r and/or -4u; G2d-lk, -2v; 
Gpi-3u; Idh-le, -2% -3m; Lap-lf; Mdh-2e and/or -2i, -3m and/or -3q; Pgm-5o, -6i 



70 DECKER-WALTERS, WALTERS, COWAN & SMITH Vol. 13, No. 1 



and Skd-li and/or -2m. Aat-3v, G2d-li, Idh-3o, and Lap-lf" are relatively rare alleles 
found in some populations. 

Distribution and ecology. Populations of var. ozarkana have been discovered in 
western Illinois, southern Missouri, Arkansas, eastern Oklahoma, and Louisiana. 



may include similar 



streams 



Mountains 



rally disturbed areas in riverine floodplains and can invade humanly disturbed 
habitats as well. Fruits are water-dispersed. This variety is distinguished from 
other varieties of ssp. ovifera by the relatively quick seed germination (usually 
within two days). The distribution and ecology of var. ozarkana are described more 
fully in Smith etal. (1992). 



:yme and m 
extremes oi 



Alabama 



We 



Alabama 



Evolutionary trends in C. pepo. 



Wil 



: /< 



derived from genetically divergent population lineages, it was relatively easy to 
detect and assess as introgressants in wild populations of ssp. ovifera those alleles 
that were otherwise restricted to and characterized ssp. pepo. First-generation 



from cultivars to wMp 



Wilson 



1987). Although recent introgression is suggested by various other populations, 
the significance and temporal depth of hybridization between cultivars and wild 
populations is unclear. Obviously, introgression into the wild has not occurred to 
the extent of obliterating a more ancient genetic background in var. ozarkana. 



voiutionary relationships among taxa of C. pepo an 
Most, if not all, primitive cultivars of var. ovifera ap; 
r. ozarkana. Variety ozarkana possesses the isozyme 

isozyme alleles in the crop (i.e., var. ouift 



wild 



subset of those in the progenitor (i.e., var. ozarkana) and the two taxa are genet 



similar 



r (ci. Doebley 



isozyme 



and their putative progenitors yielded the "subset" criterion says the following 



isozyme 



in intervention. This is why it was concluded pre 
domesticated members of ssp. pepo and ssp. ovifen 



were selected from genetically diverged wild populations already exhibiting 
distinct alleles found in these subspecies today. 

Although ssp. fraterna has a genetic identity value with ssp. ovifera var. ov 
that is also close to 0.90, ssp. fraterna apparently lacks at least three alleles {hAdh-le, 
Mdh-3m, Skd-lm) that are in high frequencies in cultivars of ssp. ovifera; hence ssp. 
fraterna fails the first criterion wherein allelps of iHp rrnn are a subset of those in the 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 71 



wild progenitor. Nevertheless, the principal component analyses did indicate 
close overall genetic affinities between a few cultivars that have been classified 
under ssp. ovifera var. ovifera (i.e., 'Flat Striped') and ssp. pepo (i.e., 'Orange Ball'), 
indicating that at least some cultivars may have been selected from populations in 
northeastern Mexico (cf. Andres 1987). Further study of ssp. fratcrna is needed to 
make these determinations. 

Most analyses revealed a relatively distinct isozyme pattern for ssp. ovifera var. 
texana. Like ssp. fraterna, Texas populations lack various alleles common in ssp. 
ovifera var. ovifera. In addition, almost all populations of var. texana contain an allele 
(Idh-3o) not found in any cultivars. Consequently, evidence is lacking which would 
suggest that any cultivar or group of cultivars have their direct origins in Texas 
populations. 

Applying Doebley's (1989) criteria in search of the wild ancestor of Mexican 
landraces of ssp. pepo, we do not find a good candidate for progenitor among 
currently known wild populations. Consequently, we believe that populations of 
this wild progenitor are undiscovered or probably extinct. Furthermore, the pat 
tern of genetic differentiation (as discussed below) between ssp. pepo and the other 
taxa, including the Mexican ssp. fraterna, suggests that the progenitor of ssp. pepo 
evolved in a distinct area of Mexico. 

Examination of C. argyrosperma ssp. sororia gave us an additional evolutionary 
perspective on isozyme variation of C. pepo. The fact that C. argyrosperma ssp. 
sororia exhibited more or less equally divergent but genetically distinct relation- 
ships with each taxon of C. pepo indicates that long periods of reproductive isola- 
tion due to geographic separation probably account for differences among ssp. 
ovifera var. ozarkana, var. texana, ssp. fraterna, and the hypothesized progenitor of 
ssp. pepo. Such a conclusion fits Wiley's model of vicariance speciation in which 
geographically disjunct progenitor populations are replaced by more divergent and 
genetically different populations through time (Wiley 1981; Wiley and Mayden 1985). 
This means that none of the wild populations known today is genetically equivalent 



pepo. This 



presuming 



mixture of other factors (e.g., selection) produced this pattern of uniform diver- 
gence among taxa of C. pepo. Of course, what requires the fewest assumptions does 
not necessarily reflect reality and factors such as human and natural selection (e.g., 
seed germination rates) and population bottlenecks (e.g., relative genetic homoge- 



rifera i 
With 



pepo. 



Mexico 



Newsom 



before at least two independent domestications took place. 



ACKNOWLEDGEMENTS 



Westheimer 



i nis study was funded Dy tne westneimer ramny ruwiunuwu, ...««,. «*-»- — .. 

the Cincinnati Museum of Natural History, the National Museum of Natural History's 
Research Opportunities Fund, administered by Stan Shetler, and Fairchild Tropical Garden. 



72 



DECKER-WALTERS, WALTERS, COWAN & SMITH 



Vol. 13, No. 1 



LITERATURE CITED 



ANDRES, THOMAS C. 1987. Cucurbita 
fraterna, the closest wild relative and 
progenitor of C. pepo. Cucurbit Genet- 
ics Cooperative Report 10:69-71. 

COWAN, C. WESLEY and BRUCE D. 
SMITH. 1993. New perspectives on a 
wild gourd in eastern North America. 
Journal of Ethnobiology 13:17-54. 

DECKER, DEENA S. 1985. Numerical an- 
alysis of allozyme variation in Cucur- 
bita pepo. Economic Botany 39:300-309. 

. .. 1986. A biosystematic study 

of Cucurbita pepo. Unpublished Ph.D. 
dissertation, Department of Biology, 
Texas A & M University, College Station. 

1988. Origin(s), evolution, 

and systematics of Cucurbita pepo (Cu- 
curbitaceae). Economic Botany 42:4- 



15. 



NEWSOM 



Numerical analysis of archaeological 
Cucurbita seeds from Hontoon Island, 
Florida. Journal of Ethnobiology 8:35- 
44. 

DECKER, DEENA S. and HUGH D 
WILSON. 1987 Allozyme variation in 
the Cucurbita pepo complex: C. pepo var 
ovifera vs. C. texana. Systematic Botany 
12:263-273. 

DECKER- WALTERS, DEENA S. 1990. 
Evidence for multiple domestications 
of Cucurbita pepo. Pp. 96-101 in Biology 
and Utilization of the Cucurbitaceae. 
David M. Bates, Richard W. Robinson, 
and Charles Jeffrey (editors). Cornell 
University Press, Ithaca, New York. 

T.W. WALTERS, U. POS- 

LUSZNY, and P. G. KEVAN. 1990 



IGNART, F. and N. F. WEEDEN 1984. Al- 
lozyme variation in cultivars of Cucur- 
bita pepo L. Euphytica 33:779-785. 

KIRKPATRICK, KURT J., DEENA S. 
DECKER, and HUGH D. WILSON. 
1985. Allozyme differentiation in the 
Cucurbita pepo complex: C. pepo var. 
medullosa vs. C. texana. Economic Bot- 
any 39:289-299. 

NEI, M. 1972. Genetic distance between 
populations. American Naturalist 106: 
283-292. 

NEWSOM, LEE A., S. DAVID WEBB, and 
JAMES S. DUNBAR. 1993. History 
and geographic distribution of Cucur- 
bita pepo gourds in Florida. Journal of 
Ethnobiology 13:75-97. 

SAS INSTITUTE INC. 1985. SAS User's 
Guide: Statistics. Version 5 edition. 
SAS Institute Inc., Cary North Caro- 
lina. 

SMITH, BRUCE D, C. WESLEY COWAN, 
and MICHAEL P. HOFFMAN. 1992. Is 
it an indigene or a foreigner? Pp. 67- 
100 in Bruce D. Smith: Rivers of Change: 
Essays on the Origins of Agriculture in 
Eastern North America. Smithsonian 
Institution Press, Washington, D. C. 

WILEY, E.O. 1981. Phylogenetics: The 
Theory and Practice of Phylogenetic 
Systematics. John Wiley and Sons, 

New York . 

and RICHARD L. MAYDEN. 

1985. Species and speciation in phy- 
logenetic systematics, with examples 
from the North American fish fauna. 
Annals of the Missouri Botanical Gar- 
den 72:596-635. 



Genealogy and gene flow among an- WILSON, HUGH D. 1989. Discordant pat 



nual domesticated species of Cucur- 



burnal 



789. 



DOEBLEY, JOHN. 1989. Isozymic evidence 
and the evolution of crop plants. Pp. 
165-191 in Isozymes in Plant Biology. 
Douglas E. Soltis and Pamela S. Soltis 
(editors). Dioscorides Press, Portland, 
Oregon. 



terns of allozyme and morphological 
variation in Mexican Cucurbita. Sys- 
tematic Botany 14:612-623. 

, JOHN DOEBLEY, and MEL- 



VIN DUVALL. 1992. Chloroplast DNA 
diversity among wild and cultivated 
members of Cucurbita (Cucurbitaceae). 
Theoretical and Applied Genetics 84: 
859-865. 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 73 



BOOK REVIEW 



Matson 



$60 



0-8165-1196-9. 



This is a well written, organized, researched and documented book, based on 
the author's own work and reports and publications of others. It is profusely 
illustrated and there are eight tables of data. I wish I could offer as much praise for 
the index, which I found disappointing in that it is largely a listing of names of 
investigators and archaeological sites and locations. Beans and squash, both culti- 
vated, are mentioned in a few places in the text, but neither is in the index under 
either common or generic names. 

As a general ethnobotanist with or.ly some exposure to archaeological ethno- 
botany, I was unfamiliar with the author and his research. Given the title of the 
book, imagine my dismay when, upon glancing through it, I discovered that most 
of the illustrations related to archaeology: diagrams of excavations, locations, and 
projectile points. My immediate reaction was to question the relevance of what 
appeared to be a superfluidity of archaeological data, discussion, argument, and 
interpretation. Although I have now read the book and see the connection, I must 
alert readers of this journal that they are likely to find the perspective on the 



dissimilar 



stemming from 



After three chapters, covering over 200 pages, (Introduction; The nature of 



Basketmaker 



The 



maize 



Matson also discusses and "tests" these models against the avail- 



from 



material 



is well integrated with that in earlier and later sections. 



from 



plants would be given recognition. But the facts of life are that these other species 
are often only poorly represented in archaeological sites (e.g. seeds of beans do 



m 



for construction, to say nothing of "testing." 



While 



mislead 



familiarity 



treatment was interesting and valuable to me and I recommend Matson 's book to 
those who are working in southwestern ethnobotany, even to those whose studies 
do not focus primarily upon looking back through time. 

Willard Van Asdall, Past Editor 
Journal of Ethnobiology 
4479 N. Summer Set Loop 
Tucson, AZ 85715, USA 



Advertising Information 



Journal of Ethnobiology 



published by the Society of Ethnobiology 

Mailing Instructions. All initial advertising contracts and correspondence 

should be sent to: 



Secretary /Treasurer 
Society of Ethnobiology 
Brien A. Meilleur 
Amy B.H. Greenwell Ethnobotanical Garden 

Bishop Museum 

P.O. Box 1053 
Captain Cook, HI 96704 

phone: (808) 323-3318 
FAX: (808) 323-2394 



Insertion orders and camera ready copy should be sent to: 



Editor, Journal of Ethnobiology 
Dr. Deborah Pearsall 
American Archaeology Division 

103 Swallow Hall 

University of Missouri 

Columbia, MO 65211 

phone: (314) 882-3038 

FAX (314) 882-9410 



— 



/. EthnobioL 13(l):75-97 



Summer 1W3 



HISTORY AND GEOGRAPHIC DISTRIBUTION OF 

Cucurbita pepo GOURDS IN FLORIDA 



LEE A. NEWSOM 
S. DAVID WEBB 

Florida Museum of Natural History 

Museum Road 
Gainesville, FL 32611 



JAMES S. DUNBAR 
Florida Bureau of Archaeological Research 

R.A. Gray Building 
Tallahassee, FL 32399-0250 



ABSTRACT.— Page-Ladson, a Florida wet site, has yielded Cucurbita pepo gourd 
seeds that date to the Pleistocene-Holocene transition. One sample with gourd 
seeds has accelerator radiocarbon dates of 12,545 ± 80 B.P. and 12,375 ± 75 B.I! 
(NSF-Arizona AMS lab, AA-7452 AA-7453). A gourd seed from a deeper stratum 
was directly dated by the accelerator method to 12,570 ± 100 B.P. (AA-8759). 
These records constitute the earliest for the genus Cucurbita from any geographic 
area. The Page-Ladson seeds and Cucurbita pepo remains from other Florida wet 
sites are compared; morphometric data from the combined collections indicate 
that all prehistoric seeds from Florida appear to be from wild gourd populations. 
The implications of recovering Pleistocene gourds in eastern North America are 
discussed, and a revised interpretation of their prehistoric distribution and cul- 
tural significance is presented. 

RESUMEN.-Page-Ladson, un sitio arqueologico pantanoso en la Florida, ha 
proporcionado semillas del calabazo Cucurbita pepo que datan de la transicion del 
pleistoceno-holoceno. Una muestra que contiene semillas de calabazo ha sido 
datada, a partir de carbono radiactivo mediante acelerador, a 12,545 ± 80 y 
12,375 ± 75 anos antes de! presente (AA-7452 y AA-7453, laboratorio AMS en 
Arizona de la Fundacion Nacional de Ciencia de los Estados Unidos de Norte- 
america). Una semilla de calabazo de un estrato mas profundo fue datada directa- 
mente mediante el metodo del acelerador a 12,570 ± 100 anos antes del presente. 
Estos constituyen los registros mas tempranos para el genero en cualquier area 
geografica. Se comparan las semillas de Page-Ladson y los restos de Cucurbita pepo 
de otros sitios cenagosos de la Florida; los datos mortometneos de las colecciones 
combinadas indican que todas las semillas prehistoricas de la Florida parecen ser 
de poblaciones silvestres de calabazos. Se discuten las impl.cac.ones de los cala- 
bazos del pleistoceno en Norteamerica oriental, y * presenta una interpretacion 
actualizada de su distribucion prehistorica y su importance cultural. 

RESUME.-Page-Ladson, un site humide de Floride, a produit des graines de 
courges Cucurbita pepo qui datent de la transition Pleistocene-Holocene. Un 
echantillon comprenant des graines de courges a produit des dates basees sur 



76 NEWSOM, WEBB & DUNBAR Vol. 13, No. 1 



l'accelerateur radiocarbone de 12,545 + 80 et 12,375 ± 75 annees avant le present 
(NSF-Arizona AMS lab, AA-7452, AA-7453). Une graine de courge provenant 
d'une couche plus profonde est datee directement par la methode d'accelerateur a 
12,570 ± 100 (AA-8759). Ce sont les donnees les plus anciennes pour le genre, 
quelque soit la region. Les graines de Page-Ladson et les restes de Cucurbita pepo 
provenant d'autres sites humides de Floride sont compares; des donnees mor- 
phometriques obtenues sur les collections combinees indiquent que toutes les 
graines prehistoriques de Floride semblent provenir de populations de courges 
sauvages. Une discussion concernant les implications presentees par la presence 
de courges datant du Pleistocene dans Test de 1'Amerique du Nord est indue, 
ainsi qu'une revision de Interpretation de leur repartition prehistorique et de 
leur valeur culturelle. 



INTRODUCTION 



preservation 



materials, including human brain tissue, woven fabric, wooden artifacts, and abun- 
dant plant structures (Coles and Coles 1989; Doran and Dickel 1988: MacDonald 



r ^v^v.v.^^i^^ V v_wx<_o nnu v-uica 1TO7, L^Uldll dllU L/lCKtri X700, 1VK1LL/VJ1 icii« 

and Purdy 1982; Purdy 1988, 1991). Gourd remains, including bottle gourd 
(Lagenaria siceraria) and Cucurbita gourd (Cucurbita pepo), have been recovered from 
waterlogged deposits at several Florida sites. Cucurbita seeds and rind are known 



from 



fears ago through the Spanish Mission Peri 
ry (Cutler 1975; Decker and Newsom 1988; 
Milanich and Fairbanks 1980:118; Mitchem 



som and Quitmye 
Newsom 



Recent excavations at the Page-Ladson (8Je591) Site in the lower Aucilla River 
add another to the list of Florida sites with Cucurbita gourd remains. Deposits at 



Wisconsin glacial maximum 



-—-©- **■ ~««.^ uv/m me wibLuiibin giaciai maximum, appruwiua^v 

18,000 years ago, to around 4000 B.P. Lithic, bone, and wooden artifacts from the 
late Paleolndian and Archaic periods appear in terminal Pleistocene, early Holo- 
cene, and more recent strata mnnhar *>* ->i ioon\ 



of 



rate excavation units at Page-Ladson. These strata also contain extinct Pleistocene 
megafauna, including horses (Equus sp.) and American mastodon (Mammut ameri- 
canum). Evidence of human activity from some of the same general strata is present, 
but scant. One sample (F.S. 131) with gourd seeds has accelerator radiocarbon 



__. ,_ j. w „. x . miu L ^ /D/D ± /D D r (isjbh-Anzona AMb laD, a/w**"-/ 

AA-7453). A Cucurbita seed from the deepest excavation level (level 26B) of Test F 



Page-Ladson. 



method 
AMS lab AA-8759). With 
been donimpntpH in ™™ 



In this paper we describe the Page-Ladson Cucurbita pepo seeds and the deposi- 
al contexts from which thev werp rpm^roH ru a p™~ t ^o™ cpoH<; are then 



compared with Cucurbita pepo 



archaeological 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 77 



We outline a reconstruction of the antiquity and geographic distribution of Cucur- 
bita pepo in Florida, incorporating the new data. We employ a biogeographic per- 
spective to explain the Pleistocene Cucurbita gourd occurrence in the eastern United 
States. Finally, we discuss the hypothesis formulated by Smith et al. (1992) and 
others that an indigenous, eastern wild Cucurbita gourd was present in North 
America as early or earlier than archaeologically-documented Mesoamerican rela- 
tives as a new way to elucidate the controversial history of gourd use in North 
America. 



PAGE-LA DSON (8JE591) 



Location and riverine environment.— The Page-Ladson site occurs below the surface of 
the Aucilla River in a section of the river known as Half-Mile Rise (Fig. 1). Like most 
of the lower Aucilla River, Half-Mile Rise consists of a string of deeper segments 
representing sediment-filled sinkholes connected by shallower links; the river rises 
from one sinkhole, courses half a mile, and then disappears into another sinkhole. 
The site is located within one of the large, partly eroded sinkhole depressions. 

Many features of the Page-Ladson site derive from its history as a coastal 
sinkhole in limestone (karst) terrain during the late glacial rise of sea level (Brooks 
1967; Vernon 1951). During the late Pleistocene, backfilling sinkholes appeared as 
deep, isolated freshwater ponds (cenotes) surrounded by well-drained limestone 
terrain. Under the more arid conditions of the late Pleistocene in Florida (Watts 
and Hansen 1988), sinkhole ponds were an important source of fresh water. 

Flowing water has removed large volumes of sediment from sinkholes in the 
river course and also has destroyed some of their defining limestone walls. For- 
tunately, the relatively deep erosion of the river channel through the sediments of 
the main sinkhole at Page-Ladson did not remove the portion along the west bank, 
where more than five meters of sediments record the late Pleistocene and early 
Holocene history of the area. 

Page-Ladson deposition and gourd -bearing contexts.-Nine excavations of various pro- 
portions have been conducted at Page-Ladson, beginning in 1983. Excavation units 
that penetrate deeply enough reach a discontinuous, late Pleistocene reddish- 
brown peat stratum. Near the center of the main Page-Ladson sinkhole we en- 
countered the oldest sediments consisting of buried peat that produced a radiocar- 
bon date of 18,430 ± 220 B.P. (Dunbar et al. 1990). The next units encountered are 
0.5-1. 0m thick calcareous sand lenses that overlie and intergrade with the red- 
brown peat. 

The red-brown peat and the associated calcarenite deposits frequently contain 
late Pleistocene vertebrates. During the most recent excavation at Page-Ladson 
(1991 Test F, Level 26B) a proboscidean (mastodon, mammoth) skull was uncov- 
ered with its tusks and lower portions buried in the red-brown peat and the cra- 
nium extending into the overlying calcareous sand. One Cucurbita pepo seed was 
found in the red-brown peat that filled the eye orbit of the skull. This particular 
gourd seed was dated by the accelerator radiocarbon method; the resulting date of 
12,570 ± 100 B.P. (AA-8750) was mentioned earlier. This seed establishes the ear- 
liest date for Cucurbita pepo at Page-Ladson. 



78 



NEWSOM, WEBB & DUNBAR 



Vol. 13, No. 1 




FIG1 



(8J 



The calcareous deposits described above are capped in several places by a 
highly organic clay stratum (Levels 16-22, 1991 unit). Some pockets of organic clay 
occur also within the calcarenites, and, similarly, as isolated pockets of material 
overlying the primary organic clay concentration of Level 20B in the Test F excava- 
tion. The organic clay deposits are coarse in texture due to an abundance of loosely 



small 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



79 



TABLE 1.— Radiocarbon dates for Page-Ladson (8Je591), Florida, 
(see also Dunbar et al. 1990) 



Age 



Test 



Sample/stratum 



3440 ± 70 B . P. 

4070 ± 60 B. P. 

8905 ± 65 B . P. 

9450 ± 100 B.P. 

9730 ± 120 B.P. 
10,000 ± 120 B.P 
10,280 + 110 B.P 
10,600 ± 70 B.P. 
10,520 ± 90 B.P. 
11,770 ± 90 B.P. 
11,790 ± 90 B.P 
12, 240 ± 90 B . P. 
12,330 ± 110 B.P. 
12,375 ± 75 B.P. 
12,545 ± 80 B.P. 
12,570 ± 200 B.P. 
12,570 ± 100 B.P. 
14,600 ± 115 B.P. 
18,430 ± 220 B.P. 



Beta 

Beta, 

Ariz.) 

Beta 

Beta 

Beta 

Beta 

Beta 

Beta 

Beta 

Beta 

Beta 

Beta 

Ariz.AMS) 

Ariz.AMS) 

Teled.) 

Ariz.AMS) 

Ariz.) 

Beta) 



B 
B 
C 
B 
B 
C 
C 
C 
A 
A 
C 
C 
B 
C 
C 
A 
F 
C 
E 



bottom of ceramic-bearing deposit 
land bridge collapse (fluviation) 
wood stake from above "Bolen floor" 
clay with mulberry wood remains 
"Bolen-beveled" horizon 
"Bolen floor" 
"Bolen floor" 

upper colluvium, calcarenite deposits 
proboscidean bone in upper calcarenite 
upper calcarenite 

calcarenite deposits 

bottom of core in unit, calcarenite deposits 

calcarenite deposit 

FS 131 1 Cephalanthus twig, woody-organic clay 

FS 131 1 Vitis seed, woody-organic clay 

upper calcarenite 

lower calcarenite/red-brown peat (Cucurbita seed) 1 

lower peat deposit 
"cypress-forest" peat 



Sample with Cucurbi fa-gourd seeds. 

fragments in one sample that has been intensively studied are strongly modal at 
7-10 mm. We suggest elsewhere (Webb and Newsom 1991) that at least some of 
this material may represent proboscidean, specifically mastodon (Mammut ameri- 
canum), dung. Two Cucurbita pepo seeds were recovered with a sample of hypoth- 
esized mastodon digesta, F.S. 131 from the 1988 excavation. Additional Cucurbita 
pepo seeds were recovered from similar deposits in Test F (Levels 14-25, see below). 
Accelerator radiocarbon dates on a buttonbush twig and a grape seed from F.S. 
131 are 12,545 ± 80 B.P. (AA-7452) and 12,375 ± 75 B.P. (AA-7753), respectively 

(Table 1). 

Remains of proboscidea and other extinct Pleistocene fauna appear in the red- 
brown peat, calcarenite, and woody organic clay deposits. Evidence at Page-Ladson 
of a direct association between early humans and Pleistocene fauna (and therefore 



mans 



limited, but not entirely absent. Worked 



manly-modified megafaunal remains have been recovered from other locations 



Webb 



recovere 



tions at Page-Ladson, along with possible early Paleolndian lithic tools. 

The issue of whether Paleolndians coexisted with Pleistocene megafauna in 
north Florida in general, and at Half-Mile Rise in particular, is beyond the scope of 
this paper. However, the earliest human use of the area is worth considering 
because the late Pleistocene/early Holocene strata at Page-Ladson include seeds of 



80 



NEWSOM, WEBB & DUNBAR 



Vol. 13, No. 1 



TABLE 2.— Page-Ladson Cucurbit a pepo contexts. 



Total 
Seeds 



Field 

No. 



Provenience 



Matrix Type 



Collection 



2 



1 

3 
8 
3 



h 



1 



3 



5 



1 



1 



1 



h 



131 



17-A 

25-A 
26-A 
26.2 



26.3 



26.4 



28.4 



29-A 



29-B 



30.4 



32-A 



1988 Test C, 
Zone D 

1991 level 14 

1991 level 21 

1991 level 22 

1991 level 22, 
southeast corner 

1991 level 22, 
southwest corner 

1991 level 22, 
northwest corner 

1991 level 23, 
northwest corner 

1991 level 24 



1991 level 24 



1991 level 25, 
northwest corner 

1991 level 26B 



unprovenienced 



] Seed in association with proboscidean skull. 



woody-organic clay, 
upper calcarenite 

woody-organic clay 

woody-organic clay 
woody-organic clay 
woody-organic clay 



woody-organic clay 



woody-organic clay 



woody-organic clay 



woody organics in 
calcarenite levels 

woody organic in 
calcarenite 

woody organics in 
calcarenite 

red-brown peat/ 
calcarenite contact 

general calcarenite 
deposits 



bulk sediment 
sample 

excavation screen 

excavation screen 

excavation screen 

bulk sediment 
sample 

bulk sediment 
sample 

bulk sediment 
sample 

bulk sediment 
sample 

excavation screen 



bulk sediment 

sample 

bulk sediment 

sample 

in situ 1 



excavation screen 



human 



- r~r~> ~ r*«*" «wi was ^ciictnuy unuzea later py numan groups m nv«w- 

and the eastern United States. The possibility exists that gourds at Page-Ladson 
are in some way tied to the earliest human appearance at the site. 

Page-Ladson Cucurbita pepo gourd seeds.— Forty-one Cucurbita pepo seeds resembling 



$ 



organic clay deposits. 



Most 



The Cucurbita pepo seeds are in excellent condition, with intact, well-preserved 

mareins and sepd mat* (V\<y l\ TUn *.*.« 1„ c t- r +*<% ~:ui^ ™* a cfnHon 



Marginal 



cases very weakly developed. The tan-brown color and smooth, even, gently curved 
margins of the Page-Ladson seeds exclude indigenous Cucurbita okeeehobeensis, the 



Mean 
Deviation 



mm 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



HI 




■ 
■ 

■ 
■ 

- 

a 


■ 




" * ' 



v * 








cm 








I IN 



Mil 



I 






b 



mm 



10 




FIG 2. 



recovered 



in association with mastodon cranium; (b) seeds from Level -- 



82 NEWSOM, WEBB & DUNBAR Vol. 13, No. 1 



TABLE 3.— Page-Ladson Cucurbita pepo seed measurements. ' 



1 lv. 14,91. 17A 



10 



19 



" 



ft 



Width 



Seed/Provenience Length Proximal Mid-sect. Distal Thickness W/L 2 



6.96 4.41 



2 lv.21,91.25A 9.48 2.49 6.28 4.42 1.04 0.66 

3 lv.22,91.26A 9.90 2.70 6.15 5.23 0.50 0.62 

4 " 10.05 3.15 7.20 4.80 0.90 0.72 

5 " 9.90 2.90 6.15 4.50 0.70 0.62 

6 " 10.40 2.45 7.10 5.15 1.20 0.68 

7 " 10.15 2.83 7.10 4.80 0.90 0.70 

8 " 10.40 2.70 6.30 4.83 1.05 0.61 

9 " 10.15 2.15 6.80 4.95 1.05 0.67 



7.10 5.00 0.90 



11 lv. 22, 91.26.2 10.46 2.88 6.80 4.51 - 0.65 

12 " 9.14 2.36 6.38 4.78 - 0.70 

13 " 10.03 2.66 6.45 4.47 - 0.64 

14 lv. 22, 91.26.3 10.15 2.65 6.72 5.07 0.55 0.66 

15 " 9.11 2.61 6.53 3.68 0.95 0.72 

16 " 10.13 2.26 6.17 4.77 1.00 0.61 

17 " 8.73 2.00 5.07 3.46 - 0.58 

18 " 9.32 2.45 6.47 4.44 - 0.69 



2.36 5.95 



20 lv. 22, 91.26.4 10.13 2.28 6.43 4.54 0.85 0.63 

21 lv. 23, 91.28.4 9.71 1.91 5.94 3.91 - 0.61 

22 " 10.03 2.65 6.71 4.48 - 0.67 

23 " 8.89 2.84 6.47 4.72 - 0.73 

24 lv.24,91.29A 9.71 2.49 6.49 4.40 0.63 0.67 
2d " 9.97 1.85 6.82 4.72 0.95 0.68 

26 " 9.61 2.15 6.86 4.87 0.79 0.71 

27 " 10.55 2.37 6.94 5.02 0.42 0.66 

28 - 2.27 5.64 - 

29 lv.24,91.29B 9.35 2.70 7 00 5 25 100 0.75 

30 lv. 25, 91.30.4 10.40 1.93 7 30 4 50 100 0.70 

31 lv. 26b, 91.32A 9.60 2.95 6.55 5.20 1.05 0.68 

32 level uncertain 9.20 3.00 7.40 5.15 0.90 0.80 

f A " 1115 3 -15 7.10 5.50 0.95 0.64 

?? " 10.50 2.95 7.60 6.40 1.05 0.72 

aUdai-J 9 ' 65 235 690 5 - 25 °- 70 ° 72 

AVERAGE 9.87 2.53 6.62 4.76 0.88 0.67 

Stand. dev. 0.54 0.35 51 53 19 0.05 

Variance 0.29 0.12 0.26 0.28 0.04 0.00 

Minimum 8.73 1.85 5 07 3 46 42 0.58 

Maximum n. 15 3#15 ?M ^ 0>80 



•Measurements in millimeters. 



Z XT 8 r T, ° n midsection Widest) width. Six fragmentary seeds wer 

2 from 1988 sample 131, 2 from lv. 21 no. 91 25A. and 2 without certain provenience. 



d: 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 83 



mid-section is 6.62 mm (S.D. = 0.51; range 5.07-7.60 mm). All specimens arc 
widest at mid-section, with the exception of a single seed from level 22 (seed 
number 3 in Table 3) that is widest just distal of the mid-section point. Overall 
length and width values for the Page-Ladson seeds fall within the range for modern 
ssp. ovifera, which includes ornamental gourds, scallop and crookneck squashes 
(var. ovifera), and wild gourds (var. ozarkana, var. texana) (Cowan and Smith, this 
volume; Decker 1986, 1988; Decker and Wilson 1986; Decker-Walters et al., this 
volume). The Page-Ladson seed dimensions, however, fall short of values for ssp. 
pepo, which regularly attain lengths of 12.0 mm and widths of at least 8.0 mm 
(Decker 1986, 1988; Decker and Wilson 1986). Width and length measurements for 
the Page-Ladson seeds compare favorably with other prehistoric assemblages of 
small-seeded Cucurbita pepo gourd from Florida (see below). 

The Page-Ladson seeds are generally rounded in shape; width to length ratios 
range from 0.58 to 0.80 (very round) (Table 3). The sample mean, 0.67, agrees with 
W/L values for ssp. ovifera (e.g., Decker 1986; Decker and Newsom 1988; Decker 
and Wilson 1986). Given their great age, we have not attempted to assign the Page- 
Ladson seeds to an extant variety. 

The coefficient of variation (C.V. = 100 x standard deviation/mean) applied 
to seed dimensions provides a more precise measure of the degree of morpholog- 
ical variability in a population than is evident using standard deviation alone. The 
C.V. for length among the Page-Ladson seeds is 5.5. This is relatively low and 
indicates a rather homogeneous population. The C.V. for width is higher, 8.0. 
High coefficients of variation may indicate presence of more than one fruit type 
and/or the initial effects of hybridization, cultivation, and selective breeding (Cowan 
and Smith, this volume). For example, values for the Phillips Spnng site (King ™»; 
Cowan and Smith, this volume:Table 4) are moderately high (C.V. length - 8.7, 
C.V. width = 9.1) and have been interpreted, along with other data, ^evidence 
that more than one form of Cucurbita gourd was present at that site, Inus, the 
somewhat smaller coefficients of variation for the Page-Ladson seeds may indicate 
that fewer, or perhaps a single, variety of gourd was present at the site. 

PREHISTORIC Cucurbita pepo GOURDS IN FLORIDA 
Page-Ladson is the eighth location in Florida from which g«*ric| V** £ 

. .. _ ... •<■ .1 ^i ^ ramuprpH IVw. 3). lne vounetr 



vvo ssp. ovif 



from 



are wet sites, primarily shell-midden 



aic wet sues, primarily biitii-iiuuucu v^ K v^~~, 

organic materials. The recovery of gourd remains in assocat.on Mrihfah^ Mtt H* 
cordage fragments suggests that at some of these sites gourds funcfoned as 

floats in a fishing tradition. t u«i««i^»llw with 

The seeds from the various Florida sites are very similar <"£**p°? *££ 
well-defined, rounded margins, sparse marginal hair, and gentle n rrowmg o d 
the seed sinus region. Occasional seeds |-«^"5SJ 
the sinus area. Thirty-two seeds from Pineland ( laWe 4) ourer y h 
dense marginal hair, an apparently tomentose seed coat and "^ *"™T 
thicker margins. These se'eds are undersized *J2^T£Z£Z?Z 



impression 



84 



NEWSOM, WEBB & DUNBAR 



Vol. 13, No. 1 



ST. JOHNS RIUER 



TICK ISL. 
DE LEON SPRINGS 

HONTOON ISL. 

GROOE' S 
ORANGE SITE 




KEV MARCO 



£f> 



# 




FIG. 3.— Florida archaeological sites with prehistoric Cucurbita pepo seeds. 



C. pepo ssp. ovifera. The distinctive Pineland seeds may represent a previous y 
unrecognized form of Cucurbita. For the present purposes, the unusual Pinelan 
seeds are classified as C. pepo ssp. ovifera with the rest of the Pineland seeds, 
unless and until further analysis indicates the seeds should be treated separately- 
Summary statistics for Cucurbita pepo gourd seeds from the Florida sites de- 
scribed above are shown in Table 5; measurements of Cucurbita pepo ssp. ovip 
var. ovifera seeds recovered from a sixteenth century well in St. Augustine are 
included for comparative purposes since these seeds almost certainly represen 
domesticated plants. Average seed length for the combined populations is 9.43 mn 
(S.D. = 1.13); average seed width is 6.22 mm (S.D. = 0.45) (Table 5). The coeffi- 
cients of variation for prehistoric seeds range from relatively low (around 4.U) 
moderately high (approaching 10.0). Rather homogenous populations with a nar- 
row range of fruit forms (morphotypes) may be indicated by the relatively to 
length coefficients of variation (3.0-6.5) for Hontoon Island, Key Marco, pinela "^ 
and as discussed earlier, Page-Ladson (length C. V. 5.5). The corresponding w i 
coefficients for these sites (4.0-10.0) suggest somewhat greater variability 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



8n 



TABLE 4.— Florida site contexts with Cucurbita gourds. 



Site 



Location 



Age 



Page-Ladson 
(8Je591) 

Carter Site 
(8Mr2061) 

Tick Island 



Aucilla River 
Jefferson Co. 

Oklawaha River 
Marion Co. 

St. Johns River 
Volusia Co. 



ca. 12,500 B. P. 



Webb etal. 1984. 



?early Holocene 



Densonctal. 1992 



DeLeon 
Springs 

Groves' Orange 

Midden 

(8Vo2601) 

Hontoon Island St. Johns River 

Volusia Co. 

coastal midden 
Lee County 



spring/sinkhole 
Volusia Co. 

Lake Monroe 
(St. Johns Riv) 
Volusia Co. 



5000-2000 B.P.; Jahn and Bullen 1978; Purdy 
& possibly later 1 1991; Newsom 

and Purdy 1990. 

Purdy 1991; Newsom and 
Purdy 1990. 

Russoetal. 1992. 



ca.5000B.P. 2 



4165 B.P. to 
ca. 500 B.P, 
possibly later 

3000-200 B.P 



(8Vo202) 

Pineland 
(8LL37) 



Key Marco 



St. Augustine 



mangrove swamp 
Collier Co. 

Spanish well, 
St. Augustine 



ca. 1900 B.P, 
possibly also as 
late as 500 B.P. 

1100-1300 B.P 



Decker and Newsom 1988; 
Newsom 1987; Purdy 1991. 

Marquardt 1992. 

Walker and Marquardt 1994 



16th century 



Cushing 1897; Cutler 1975; 
Gilliland 1975. 

Scarry 1985. 



'Seeds (3 total) came from a sediment core that penetrated shell midden ^^f^™** and 
St. Johns Plain ceramics series (ca. 2000 B.C.-A.D. 500 [Milanich and Fairbanks 1980:149)). 

^Seeds (3) came from a peat core at 10 cm above a dugout canoe with a radiocarbon date of 5140 ±100 
years B.P. (Beta-14893). The canoe rested on limestone bedrock and was buned m the cucurb.t-beanng 
peat; the relationship between the canoe (date) and seeds is uncertain. 

Similarities among the coefficients for Hontoon, Key Marco, Pineland, and 
Page-Ladson-the last almost certainly representing free-ranging gourds 
gest that little or no selective pressures were placed on local gourd populations by 
human inhabitants of the first three sites. In contrast, values for Groves Orange 

- '• = 14.0, width) and the length coefficient tor 



Midden 



domest 



12.0, length; C.V. 



18.3) are at least twice as 



me domesticated population irunui.nu B u^..M , rmlIOC - 

high as the values for the other Florida sites. The relatively high values for Groves 

° r • i- „*~a fk-»f ic PYrpntionallv large 



Midden 



mm, width 8.41 mm) compared to the rest of the group 
Thislarge seed, from Level 3, may represent a distinct morpho ype, . 
that at least two forms of Cucurbita pepo may have been present at th 
Orange Midden site (if this seed is excluded, the Groves Orange M 



extremely 



cients decrease to b.D tor lengrn anu y.v i« «««..,• — „ roc ~ n co of a 

length coefficient of the St. Augustine seeds probably reflects te^« of * 

more diverse array of Cucurbita pepo fruits and/or «" "Pf^ffi,^^ 
St. Aueustine seeds derive from cultivated population(s) that undoubtedly mcluded 



86 



NEWSOM, WEBB & DUNBAR 



Vol. 13, No. 1 



0) 



73 






c 

G 



CD 

gd 






a 



r o 



o 






c/3 



05 



c/} 
03 

s 



C 
01 



I 







CO 




t: 





c 

cd 




> 

U 



Q 

01 



be 

c 

03 



c 

re 

CD 




> 

■ 

u 



Q 

01 



CD 
CJD 

c 

rV 



cd 




2 



o 



-2 
ex 

a 

TJ 
CD- 
CD 

01 



ON00^N^000000TH(JfSts 



IN 

I 
IN 

00 



(N 



LO 



I 



ON 



inNio^Nincoo^ 



^fninfOlDr^Of^fN 



cd 



lo (N 

^ o 



in 

ON 



(N cn o 

o o m 



LO CN ^h 

n lo 't 



oo 

i 



IN 

I 

(N 
LN 



00 

I 

<N 
(N 



CO 

I 

CN 
CN 



IN 

I 

CM 



IN 

I 



co 



LN 

I 

00 
IN 



00 

I 

LO 
CN 



m^^o^^'tLn^Ln 



coooovo-^oloo 



LO rf (N vD CO Tf \£) 



inNNONOfN^O^ 

sDinLnNin^inxN 



O O O CN 



I 

lo 

on 



i 

LO 



I 

LO 



I 

LO 

On 



I 

LO 



I 



\0 O v£5 \D IN 



LO 

00 



I I 

NO LO 

OS IN 

IN LN 



(N 
LO 

co 

I 

on 

oo 

■ 

IN 



00 



LO 



LN 

I 
LN 



LO 



^OLONO^OCOLOLNCN^tfCNCOlNCN 
COO^fONr-iCOCOCOLOCNNOcO^C 



\D^DLNLO^OLO\O^O^D^LOLO^O 



LO 

LO 



LO 



LO 



I 

CO 
LN 

CO 



O^t-hlNCOCCO^LOLOOOLN 

lnoononcnco^colocncnlocc 

on o co on co on on on o co ln on 



inO^rtONr^XX 
<N CN ON r^ O t* On 

© On CM 



(N 



T3 

C 



CN 

i 

a. 



i 

a; 

a. 



it 



C/J "" 

3 C 

3 

< c 

ai X 



c c 

o 

o o 



eft 

I 

c 



a 



03 
C 

eft 
i 

C 


o 



c c c 
o o o 

XXX 



o 



o 

a Td 

S c 

2 <* 



CD 

c 

6-g 



a; .2 



CD 

c 



CD 

> 

C 



2 

3 



-^ 

u 



C 

CX a» 
01 *s 




LN \0 LN 

VsD \0 vO 



LN ^ LN 



LO LO CN 

r*< (N ^ 



(N vD (N 
CN rf 



vQ \D ^sD 



CN ON ON 



CO LN Tf 

CO 00 



CO CO (N 

"^ CO 



ON <Js Qs 



LO 

CO CO 

co 




U C- U H 



a> os 

Q U 



7 3 

CD < 



z 

< 

Q 
Z 
< 



o^ 
U 

en 2 

Q < 



H 



^ z 

^ < 



c 
c 



- 

> 






u 



C 

<j 

II 
> 

u 



> 

■a 



A3 

c 



LD 



D 
cn 

— 






c 

■J) 



c 

6 
2 



V5 

c 

cu 



> 



3 



cu 
cu 

In 

a- 

3 

<T3 

c 

CL 

>^ 

CU 



H 



E 

CL» 



CD 

c 

c 






a. 



a ^ 



T3 

CD 



2 

o 
to 



CD 



.2 C 



« z: 






C 

CD 



i 






^r v 

si 

aj -a 

.£ > 



en © 



3) 



D- 



CD 

cy 



CD 



i 1 « 

5 ? S 

£ c 5 

^ o 7 
•52x1 

05 



a-.S 



c 




cr. 



5- 



CD 

u 

5 x 

o 



03 
75 



C/j 



a» eft 








^D 



13 bD O 
^ C T3 

"C T3 

D C 

T3 'C 

o ra 

SI 

u C «i 

«ft. a--- 

CD E 



CD c^ 

> T3 

CD CD 

- CD 



^ 



S 



x .si cu 

eft c -c 

O ^ H3 

> 



cu --- 
T3 ^ 



^ QJ ^ 



u 

03 

00 
00 

on 



en 

I 

t/i 

£ 



£ 

CD 

z 

c 

— 
CD 

CD 

D 

-a 

CD 



cy^ 



CD 

x. 

(J) 

I 

c 

c 



CD 

— 

03 



03 

c 

3- 

CD 



03 

cr. 

CD 



c 

O 

T3 

C 

7J 





o 



(75 

CD 

CD 

— 

03 

(N 

i 

CD 



I 



C 

c 

X 

£ 

c 



SL 



O 

"ill 
03 

£ 

-o 

CD 
O 

g 

C 

"O 

-o 

1 

CD 

— 

c 



'c/5 

03 



T3 

D 

"O 
C 



x 

r, 



C 

03 

c 

03 

TD 

CD 
N 

*c7 



en 

- 

o 





c 

CD 



C/i 

CD 



CD 



oo 



s 

c 

C/5 
C 

c 



OS 

3 

o- 



y 

— 



CD 

c- 



Summer 1993 JOURNAL 



K7 



domesticated Cucurbito pepo, and a gardening situation where hybridization be- 
tween introduced and native forms of Cucurbito was unimpeded by distance or 
isolation. 

For the purposes of this analysis the St. Augustine seeds serve as an archaeo- 
logical example of domesticated Cucurbito pepo gourd, while Page-Ladson, at the 
other extreme, is representative of wild, free-ranging Cucurbito. With these stan- 
dards, the coefficients of variation may be used indirectly to verify the advent ot 
gardening or the intensification of gardening practices and perhaps human-direc- 
ted selection for fruit characters at each of the Florida sites (Decker and News. im 
1988:40-42; also see Heiser 1989). By this measure, there is an increase in the 
length and width coefficients from 3.4 and 5.0, respectively (Table 5), for the pre- 
historic Hontoon Island snail-shell midden to 6.6 and 7.0, respectively, for the 
primarily mussel-shell midden that accreted during the latest prehistoric-historic 



may 



m 



// 



Newsom 



ition of local gourd populations. The 
type-2" (scallop-type) Cucurbito pep 



Th 



and ranges (length 10.25-11.57; width 6.72-797) for the Hontoon Island type-2 
seed sample are, however, within the limits of seed si/e variability for free-ranging 
gourds as documented by Cowan and Smith (this volume). 



FLORIDA GOURD POPULATIONS IN BROADER PERSPECTIVE 



It is probable that Cucurbita pepo seeds from St. Augustine, and possibly also 



M 



The 



rem 



seed size, and corroborate other morphological characteristics associated with 
independent or free-ranging Cucurbita pepo gourds. This baseline is important 
because it provides criteria of seed size and rind character that may be employed in 
determining which prehistoric gourd populations in eastern North America rep- 
resent free-ranging and which represent truly domesticated forms of Cucurbita 
pepo, whether introduced or developed in situ. Such determinations are necessary 
to understand the timing and trajectory of indigenous plant husbandry systems. 

The data base provided by the Florida seeds is valuable because of its time 
range— more than 12,500 years— and particularly because the Page-Ladson gourds 
are unlikely to represent domesticated plants. Thus, we have a paleontological 
example of what numerical values may be considered representative of clearly 
wild plants, as opposed to gourds under incipient or full domestication. The sheer 
quantity of gourd material from Florida is also useful in establishing a reliable base 
of figures with which to compare other data. This information may then be used in 
conjunction with data generated from modern and extant free-living gourd popu- 
lations (Cowan and Smith, this volume; Decker 1986, 1988). 

As a general rule, archaeological assemblages in which gourd seed lengths 
surpass 11.00 mm are considered to represent gourds that have undergone some 
degree of domestication (King 1985; Cowan and Smith, this volume). Individual 



88 NEWSOM, WEBB & DUNBAR Vol. 13, No. 1 



seeds with lengths greater than 11.00 mm are absent among the prehistoric Cucur- 
bita pepo gourd seed assemblages from Florida, with the exception of a single seed 
from Page-Ladson (Table 3). All seeds greater than 11.00 mm from Hontoon Island 
come from historic period deposits; the large seed from Level 3 of the Groves 
Orange Midden may also be recent, based on the proximity of historic period 
artifacts. As the 12,500 year old seeds from Page-Ladson almost certainly come 
from wild gourds, presence of a single seed larger than 11.0 mm at that site under- 
scores the cautionary note by Cowan and Smith (this volume) and others that seed 
size alone is an unreliable indicator of domesticated status. Likewise, just as 
the > 11.00 mm length baseline should be applied cautiously when attempting 
to interpret the timing and scale of domestication, so then should smaller seed 
sizes not always be equated with the wild state (Cowan and Smith, this volume; 
Decker and Newsom 1988). Interestingly, some of the smallest seeds from Hon- 
toon Island appear in the latest strata when overall Cucurbita pepo seed morpholog- 
ical diversity appears to reach its highest level (Decker and Newsom 1988). 

Thus, seed size, to the extent upon which it can be relied, generally places the 
prehistoric archaeological populations of Cucurbita pepo from Florida under the 
size range (> 11 mm) commonly associated with domesticated forms. By this 
measure, sixteenth century specimens from St. Augustine, with lengths that at- 
tain 15.00 mm, are the only seeds that fit securely within size ranges established 
for seeds of squash/gourd domesticates. 

Rind thickness, rind texture, and peduncle diameter are characteristics that 
have also been used to assess free-ranging versus domesticated status in Cucurbita 
pepo gourds. The Hontoon Island small-seeded form of Cucurbita pepo gourd 
(Table 5: Hontoon-snail, type-1) is accompanied by smooth-surfaced, thin-shelled 
rind and peduncles with relatively small basal diameters (Table 6). Mean rind 
thickness of less than 2 mm and a lack of lobing and/or wartiness are charac- 
teristics that researchers have associated with free-living Cucurbita pepo gourds 
(Cowan and Smith, this volume). Rind from the Hontoon Island collection, there- 
fore, being thin and smooth-surfaced, does not differ morphologically from that of 
wild gourd, at least in any way we can recognize. Furthermore, Cowan and Smith 
(this volume) have compared basal peduncle diameters (as a direct reflection of 
relative fruit volume) from free-living and cultivar forms of Cucurbita pepo gourd, 
concluding that peduncle diameters larger than 8 mm are more consistently asso- 
ciated with domesticated status (mean 9.75 mm, range 6.20-15.10 mm). However, 
Cowan and Smith's data show that free-living forms may approach 9 mm (mean 
5.60 mm, range 5.30-8.60 mm). The mean for Hontoon Island peduncle diameters 
(7.61 mm) is about equidistant between modern free-living and cultivar gourds, but 
the overall values are generally more similar to the free-living specimens (Table 6). 
Two peduncle diameters for Pineland are similar. Together, these characteristic 
(relatively small seeds, thin rinds, small peduncle diameter) demonstrate that the 
most abundant seed type at Hontoon Island represents a Cucurbita pepo gourd that 
was free-ranging or otherwise displayed no phenotypic alteration that is readily 
attributed to direct human manipulation. 

Whether the larger, type-2 seeds present at Hontoon Island represent a spon- 
taneous mutation from the first type of gourd, intentional selection, or a gourd 
introduction, is not clear, but there is no associated thicker (> 2 mm) or warty 



s 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 89 



TABLE 6.— Rind thickness and peduncle diameters for Florida gourd 
assemblages. 1 



Rind Thickness Peduncle Diameter 



Site N Mean Range STD N Mean Range 



Hontoon Island 30 1.08 0.72-1.55 0.22 7 8.15 7.00 9.80 

Key Marco ? - 1.80-2.20 - 



Pineland 



2 8.23 7.43-9.04 



'Measurements in millimeters. 



rind at Hontoon Island indicating a domesticated form of Cucurbito. Cutler 



Mar 



Marco 



all have smooth rind surfaces, however, like those from Hontoon Island. 

Numerical analysis (Decker and Newsom 1988) of a sample of 252 Cucurbita 
pepo gourd seeds from Hontoon Island classifies all specimens within ssp. ovitera. 
Moreover, the small seed type from Hontoon Island exhibits a strong affinity (64% 
classified) with wild var. texana, while larger seeds, including the type-2 scallop 
specimens, fall within var. ovifera. The strong identification with var. texana further 
supports the general impression that Hontoon Island gourds, especially those 
from prehistoric levels that predate the appearance of the larger-seeded type, 
were not advanced in the domestication process or at least do not display charac- 
teristics generally associated with domestication. The same may be true of the 
other Florida sites for which the majority of seeds are less than 11 mm in length, 
but currently there are few or no rinds or peduncles to corroborate the evaluation 
of domesticated versus free-ranging status utilizing seed length. Moreover, the 
fact that plant cultivation was of little or no significance throughout most of Flor- 
ida's prehistory (Milanich 1987; Newsom 1986; 1991, 1993; Scarry and Newsom 
1992), further supports the conclusion that prehistoric people in Florida typically 



m 



was exerted on local gourd populations that is not detected by the measures de- 



scribed above. 



It is worth noting that the abundant, small-seeded morphotype from Hontoon 
Island occurs throughout the stratigraphic sequence at that site and may have 



surviv 



survive 



sive, evidence of its nondomesticated status. Cucurbita pepo gourd remains from 
Hontoon Island provide the longest, continuous record in Florida for the gourd's 
presence, extending from a prehistoric deposit of ca. 3000 B.P. to as late as the mid- 
eighteenth century, based on radiocarbon dates and artifacts of European origin. 

The absence of free-ranging gourds along the St. Johns River today is an 
interesting problem 



Will 






^rowing in the St. Johns River basin (Harper 1958). Although his description may 
applv to the endemic Okeechobee gourd (Cucurbita okeechobeensts) (e.g., Walters 
and Decker-Walters 1993; Martin 1992), it is possible that what Bartram saw were 



90 NEVVSOM, WEBB & DUNBAR Vol. 13, No. 1 



descendants of the archaeological St. Johns River basin gourds. Bartrum's descrip- 
tion of "the wild squash climbing over lofty limbs of the trees; their yellow fruit 
somewhat of the size and figure of a large orange, pendant from the extremities of 
the limbs over the water" [emphasis added] (Harper 1958:87) possibly applies to 
Cucurbita pepo gourds. Fruit color (yellowish to various shades of tan/brown) and 
shape (based on the ca. 15-degree angle of curvature) of rind specimens from 
Hontoon Island are consistent with Bartram's description of a yellow (?dried), 
roundish, orange-sized fruit. Moreover, 1774 is not far removed from the terminal 
radiocarbon dates for Hontoon Island: 170 + 50 B.R (A.D. 1770), 220 ± 45 B.P. 
(A.D. 1730), and 260 ± 50 B.P. (A.D. 1680) (Purdy 1987, 1991). The late survival of 
Cucurbita pepo gourds at Hontoon Island, together with Bartram's description of 
gourds growing somewhere in the vicinity of Lake Dexter, located in the middle 
river basin only 12 miles north of Hontoon Island, may attest to survival of the 
archaeological gourd population into relatively recent times. Bartram related to 
Muhlenberg (Harper 1958:633) that hunters called the gourd "wild squash/' The 
name, coupled with Bartram's description of the habit, implies that the gourds 
retained characteristics and behaviour associated with the wild state— specifically 
the ability to grow and disperse spontaneously, without human intervention. Bar- 
tram's portrayal of gourds hanging directly above the river provides a glimpse of 
how readily fruits could have dispersed along the water course, and supports the 
suggestion of a long-established pattern of natural dispersal by water routes, as 
Smith et al. (1992) have hypothesized regarding the niche and expansion of "east- 
ernized" Cucurbita pepo gourds. 



AMERICA 



Middle 



Sm 



Watso 



esized 3000-4000 B.P. time frame when gardening and human-induced morpho- 
logical change in gourd specimens are thought to appear (Smith et al. 1992). 
Previous to the Page-Ladson discoveries, the earliest (7000 B.P.) remains possibly 
attributable to Cucurbita pepo were from west-central Illinois (Conard et al. 1984; 
Cowan and Smith, this volume); dates nearly as early come from eastern Ten- 
nessee (6990-5300 B.P.) (Crites 1991). Dates of around 4500 B.P. are recorded for 



Missouri 



from 



Maine 



method to 5695 ± 100 B.P. (AA-7491) (James 



communication, 1992). The Florida Cucurbita pepo gourd identifications thus extend 
the geographic and temporal records for early gourds. The 12,500 B.P. radiocarbon 
dates for Page-Ladson push the presence of eastern gourd back into the closing 
stages of the Pleistocene epoch. 

Some previous interpreters of early Cucurbita /*>;;o-like gourds in the eastern 

1 I t t i * * « * i 4 -^ - ^ 



Mexico 



via down-the-line exchange or similar mechanisms (e.g., discussion summarized 
in King 1985). Some researchers consider the earliest appearances in the East of 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 91 



Cucurbita pepo gourd to represent fully-domesticated forms (Asch and Asch 1992; 
Kirkpatrick and Wilson 1988; cf. Fritz 1990). The Page-Ladson seeds document 
that an eastern Pleistocene Cucurbita pepo gourd may have been growing in Florida 
prior to the appearance of Paleolndians, and thus strengthen the case for introduc- 
tion by natural, rather than by cultural means, as well as presenting the possibility 
for entry into North America prior to 12,500 B.P. 

It is not difficult to envision how Pleistocene Cucurbita pepo gourds could have 
dispersed from an early Mexican origin and center of diversity (Decker- Walters el 
al., this volume; Nee 1990) to Florida and the eastern United States without the aid 
of human intervention. Indeed, Decker-Walters et al. (this volume) postulate an 
ancient range for wild Cucurbita pepo that stretched from central or southern Mex- 
ico north to the Ozark Plateau and east to the Florida peninsula. Furthermore, the 
coastal plain and Florida platform were considerably larger in the late Pleistocene, 
when much lower sea levels exposed vast areas of the now-submerged continental 
shelf stretching from Florida to Yucatan. The greatly expanded coastal plain, known 
to paleogeographers as the Gulf Coastal Corridor, became a major migration route 
for Pleistocene mammals (Morgan 1991; Webb 1989, 1991; Webb and Wilkins 1984), 
as well as for terrestrial vegetation (Long 1974; Tomlinson 1980). Some 17-34% of 
successive Pleistocene-neotropical mammalian faunas dispersed through this 
corridor and became established in the eastern United States (Webb and Wilkins 
1984). That proboscideans may have acted as animal vectors for early gourd dis- 
persal has been suggested by other researchers (Janzen and Martin 1982; Nabhan 
1987); the possibility that one context for the Page-Ladson seeds is mastodon 
digesta is especially intriguing in this regard. 

Mirroring and perhaps in conjunction with Decker-Walters's (Decker- Walters 
et al., this volume) hypothetical range expansion of an early Cucurbita pepo ances- 
tor, we postulate a natural eastward dispersal of wild Cucurbita pepo around the 
Gulf Coastal Corridor from source areas in Mexico. Heiser (1979, 1989, 1990) and 
others have discussed how early bottle gourd (Lagenaria siceraria) might have first 
appeared in Florida (7290 ± 120 B.P [Doran et al. 1990]) and the eastern United 
States by water-dispersal along the Gulf Coast from tropical regions. Given the 
apparently similar adaptation of free-ranging Cucurbita pepo (Cowan and Smith, 
this volume; Smith et al. 1992), we envision, as does Smith (1992:285), the small 
round gourds being deposited at the mouths of rivers along the Mexican coast and 
subsequently being transported eastward along the coast by drift and the clock- 
wise flow of the Gulf Stream (Gunn and Dennis 1976; Guppy 1917), to end up at 
the mouths of other river systems from present-day Texas to Florida. Indeed, a 
look at distribution maps constructed by Nee (1990:Fig. 2) for wild var. texana and 
other closely related types reveals a strong association with Gulf coastal regions. 
Once established in the lower basins of easterly rivers, the wild gourds could 
eventually "migrate" with annual floods, and perhaps also with the aid of large 
mammalian dispersers, until their populations reached well up into the mid- 
western United States, culminating in the long-established "easternized" gourd 
of Smith et al. (1992). The process of natural dispersal and movement of gourds 
into and around the greater eastern temperate region may have been ongoing for 
millennia, and eventually aided by human groups who took up the gourds as a 
useful and edible plant. 



92 NEWSOM, WEBB & DUNBAR Vol. 13, No. 1 



The Gulf coast and Florida were cooler than present in the Pleistocene, enough 
so that at the glacial maximum of ca. 18,000 years B. P. spruce pollen is recorded for 
northern Florida (Watts and Hansen 1988; Watts et al. 1992). At other times, such 
as following the peak glacial advance of 18,000 B.P, the Gulf coast was consider- 
ably more arid than present, but ambient temperatures were not appreciably lower 
(Watts and Hansen 1988). Cooler conditions along the Gulf coastal plain may have 
allowed the tropically-derived Cucurbitaceae to adjust to climatic conditions along 
the routes of entry into more temperate zones, again facilitating the naturalization 
("easternization") of gourd populations. 



CONCLUSIONS 



At least as early as the terminal Pleistocene, Citcurbita pepo gourds dispersed 



Mexico (Decker- Walte 



Florida and probably elsewhere in the eastern United States. New evidence from 
the Aucilla River places Cucurbita pepo in Florida by 12,500 years ago. The status of 
early gourd remains in North America has been controversial. The question has 
been whether Cucurbita pepo gourds that pre-date clearly domesticated forms of 



Mexico 



America 



curbita gourds were an indigenous wild type or types that were long utilized and 
eventually taken into cultivation and domesticated by native peoples of the east- 
ern woodlands (Fritz 1990; Smith 1987; Watson 1989). The 12,500 year old Cucur- 
bita pepo seeds from Page-Ladson arguably pre-date domestication of the species, 
although the 8000-10,000 B.P. reportedly domesticated specimens from Mexico 
closely approximate this age (Whitaker et al. 1957; Whitaker and Cutler 1971; cf. 
Flannery 1973, and Heiser 1989, 1990). The basis for the domesticated status of 
early Mexican Cucurbita remains is unclear, however, and in light of the early 
material from Florida, is greatly in need of reevaluation. Unless we are ready to 
consider the possibility of Paleolndian gardening and plant protection, it seems 
most plausible to assume that the earliest eastern gourd finds are remains of wild 
plants. 

The growing archaeological record of earlier and more widely distributed Cu- 
curbita pepo gourd remains in the East, however, lends increasingly greater sup- 
port to the interpretation that the plant was present as part of the native flora by 



Walters 



Walters 



amon 



pepo f modern free-living gourds in the east, and other wild gourds, leading her 



Walters 



ment 



Numerical analysis of seeds from Hontoon Island, Florida (Decker and Newsom 



it is 



■ ^f w w 

a representative wild form, var. tcxana, and archaeological specimens. Thus, it i 
becoming increasingly more difficult to explain eastern Cucurbita pepo and indige 
nous plant production systems as a function of, or derivative of, borrowed plants 



Mexico 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



93 



Smith 



Watson 1989; Wh 



and Carter 1946) that eastern Cucurbita pepo gourds are part of the native flora, and 
that within the sphere of early plant domestication focused on indigenous wild 
taxa, a separate, eastern United States center of domestication existed for Cucur- 
bita pepo. 



ACKNOWLEDGEMENTS 



Research on the plant remains from the Page-Ladson site was partially supported by a 
grant from the National Geographic Society. We are grateful to John Ladson and family for 
allowing us to carry out our work at the site and nearby areas. We appreciate the space and 
other support provided by the Florida Museum of Natural History. Seeds from Hon toon 
Island, Tick Island, DeLeon Springs, and Groves' Orange Midden were analyzed as part of 
wet site research under Barbara A. Purdy's initiative, also partly supported by the National 
Geographic Society. William H. Marquardt and Karen Jo Walker, both of the Florida Mu- 
seum of Natural History, made the archaeobotanical research on the Pineland site possible, 
and William Marquardt provided access to the Museum's Key Marco accessions. Robin 
Denson and Bill Marquardt are responsible for the Oklawaha River Survey that recently 
resulted in the gourd identification from the Carter Site. Thanks also to Margie Scarry who 
made the St. Augustine seeds available for analysis. Special thanks to Wes Cowan, Richard 
Ford, Gayle Fritz, Fran King, Donna Ruhl, Bruce Smith, Deena and Terrance Walters, Patty 
Jo Watson, and Elizabeth Wing for their comments on earlier drafts of this paper. Finally 
many thanks to Mike Stallings for his cartographic skills and for keeping a watchful eye out 
for especially tempting peat deposits. 



LITERATURE CITED 



ASCH, DAVID and NANCY ASCH. 1992. 
Archaeobotany Pp. 177-293 in Geo- 
archaeology of the Ambrose Flick Site. 
Russell Stafford (editor). Kampsville 
Archaeological Research Center, Re- 
search Series 10, Kampsville, Illinois. 

BROOKS, H. KELLY. 1967. Rate of Solu- 
tion of Limestone in the Karst Terrain 
of Florida. Florida Water Resources Re- 
search Center, Publication No. 6, Talla- 
hassee. 

COLES, JOHN and BRYONY COLES. 
1989. People of the Wetlands: Bogs, 
Bodies, and Lake-dwellers. Thames 
and Hudson, New York. 

CONARD, N., D.L. ASCH, N.B. ASCH, 
D. ELMORE, H.E. GOVE, M. RUBIN, 
J. A. BROWN, M. D. WIANT, K. B. 
FARNSWORTH, and T. G. COOK. 
1984. Accelerator radiocarbon dating 
of evidence for prehistoric horticulture 



at the 47th Annual Meeting of the 
Southeastern Archaeological Confer- 
ence, Mobile, Alabama. 
^__ and BRUCE D. SMITH. 1993. 



443 



WESLEY 



cucurbits from the Cumberland Plateau 
of eastern Kentucky. Paper presented 



New perspectives on a wild gourd in 
eastern North America. Journal of Eth- 
nobiology 13:17-54. 

CRITES, GARY D. 1991. Investigations 
into early plant domesticates and food 
production in Middle Tennessee: A 
status report. Tennessee Anthropolo- 
gist 16:69-87 

CUSHING, FRANK H. 1897 Exploration 
of ancient key-dweller remains on the 
Gulf Coast of Florida. Proceedings of 
the American Philosophical Society 25 
(153), Philadelphia. 

CUTLER, HUGH C. 1975. Two kinds of 
gourds from Key Marco. Pp. 255-256 
in Marion S. Gilliland: The Material 
Culture of Key Marco, Florida. Univer- 
sity Presses of Florida, Gainesville. 

DECKER, DEENA S. 1986. A Biosystem- 
atic Study of Cucurbita pepo. Unpub- 
lished Ph.D. Dissertation, Department 



94 



NEWSOM, WEBB & DUNBAR 



Vol. 13, No. 1 



of Biology, Texas A&M University, Col- 
lege Station (University Microfilms, 
Ann Arbor). 

. 1987. Allozyme variation in 

the Cucurbita pepo complex: C. pepo var. 
ovifera vs. C. texana. Systematic Botany 
12:263-273. 

. 1988. Origin(s), evolution, 

and systematics of Cucurbita pepo (Cu- 
curbitaceae). Economic Botany 42:3-15. 
and LEE A. NEWSOM. 1988. 



Numerical analysis of archaeological 
Cucurbita pepo seeds from Hontoon Is- 
land, Florida. Journal of Ethnobiology 
8:35-44. 

DECKER, DEENA S. and HUGH D. 
WILSON. 1986. Numerical analysis of 
seed morphology in Cucurbita pepo. 
Systematic Botany 11:595-607. 

DECKER-WALTERS, DEENA S v TER- 



RENCE WALTERS 



C. WESLEY 



COWAN, and BRUCE D. SMITH. 
1993. Isozymic characterization of wild 
populations of Cucurbita pepo. Journal 
of Ethnobiology 13:55-72. 

DENSON, ROBIN L., LEE NEWSOM, 
BRINNEN CARTER, R. KUEHL, and 
S. CROWNOVER. 1992. Final report 
of the Oklawaha River survey. Report 
on file, Florida Bureau of Archae- 
ological Research, Tallahassee. 

DORAN, GLEN H. and DAVID N. 
DICKEL. 1988. Multidisciplinary in- 



Windover 



Wet 



bara A. Purdy (editor). Telford Press, 

Caldwell, New Jersey. 

and LEE A. NEWSOM. 1990. 



A 7,290-year-old bottle gourd from the 
Windover Site, Florida. American 
Antiquity 55:354-360. 

DUNBAR, JAMES S., S. DAVID WEBB, 
and DAN CRING. 1990. Culturally and 
naturally modified bones from a Pa- 
leoindian site in the Aucilla River, 
North Florida. Pp. 473-497 in Bone 
Modification. Rob Bonnichsen (edi- 
tor). University of Maine at Orono 
Press, Orono. 

FLANNERY, KENT V 1973. The origins of 
agriculture. Annual Reviews of An- 
thropology 2:271-310. 

FRITZ, GAYLEJ. 1990. Multiple pathways 
to farming in precontact eastern North 
America. Journal of World Prehistory 
4:387-435. 



GILLILAND, MARION S. 1975. The Ma- 
terial Culture of Key Marco, Florida. 
University Presses of Florida, Gaines- 
ville. 

GUNN, CHARLES R. and J.V DENNIS. 
1976. World Guide to Tropical Drift 
Seeds and Fruits. Quadrangle/The 
New York Times Book Company, New 
York. 

GUPPY, H.B. 1917. Plants, Seeds, and Cur- 
rents in the West Indies and Azores. 
Williams and Northgate, London. 

HARPER, FRANCIS (editor). 1958. The 
Travels of William Bartram; Natu- 
ralist's Edition. Yale University Press, 

New Haven . 
HEISER, CHARLES B. 1979. The Gourd 
Book. University of Oklahoma Press, 

Norman. 

. 1989. Domestication of Cu- 

curbitaceae: Cucurbita and Lagenaria. 
Pp. 472-480 in Foraging and Farming: 
The Evolution of Plant Exploitation. 
David R. Harris and Gordon C. Hill- 
man (editors). Unwin Hyman, London. 

. 1990. New perspectives on 



the origin and evolution of New World 
domesticated plants. Summary, in New 
Perspectives on the Origin and Evo- 
lution of New World Domesticated 
Plants. Peter K. Bretting (editor). Eco- 
nomic Botany 44(3 SUPPLEMENT): 

111-116. 

JAHN, OTTO L. and RIPLEY P. BULLEN. 
1978. The Tick Island site, St. Johns 
River, Florida. The Florida Anthro- 
pologist 31(4), part 2:1-25. 

JANZEN, DANIEL H. and PAUL S. MAR- 
TIN. 1982. Neotropical anachronisms: 
The fruits the gomphotheres ate. Sci- 
ence 215:19-27. 

KAY, MARVIN, FRANCIS B. KING, and 
CHRISTINE K. ROBINSON. 1980. 
Cucurbits from Phillips Spring: New 
evidence and interpretations. Ameri- 
can Antiquity 45:806-822. 

KING, FRANCIS B. 1985. Early cultivated 
cucurbits in eastern North America. 
Pp. 73-98 in Prehistoric Food P r0 ° uC " 
tion in North America. Richard I. Ford 
(editor). Anthropological Papers 75, 
Museum of Anthropology, University 
of Michigan, Ann Arbor. 

KIRKPATRICK, KURT J. and HUGH v. 
WILSON. 1988. Interspecific gene 
flow in Cucurbita: C. texana vs. C W ' 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



95 



American Journal of Botany 75:517- 




LONG, ROBERT W. 1974. The vegetation 
of southern Florida. Florida Scientist 
37:33-45. 

MACDONALD, GEORGE F. and BAR- 
BARA A. PURDY. 1982. Florida's wet 
sites: Where the fragile past survives. 
Early Man 4(4):4-12. 

MARQUARDT, WILLIAM H. (editor). 
1992. Culture and Environment in 
the Domain of the Calusa. Institute of 
Archaeology and Paleoenvironmental 
Studies, Monograph 1, Gainesville, 
Florida. 

MARTIN, DAVID. 1992. Endangered and 
threatened wildlife and plants: Pro- 
posed endangered status for a Florida 
plant: Okeechobee gourd. Federal Reg- 
ister 57(98):21381-21386. 

MILANICH, JERALD T. 1987. Corn and 
Calusa: DeSoto and demography. Pp. 
173-184 in Coasts, Plain, and Deserts, 
Essays in Honor of Reynold J. Ruppe. 
S.W. Gaines (editor). Arizona State 
University, Anthropological Research 
Papers, No. 38, Tempe. 
and CHARLES H. FAIR- 
BANKS. 1980. Florida Archaeology. 

Academic Press, New York. 
MITCHEM, JEFFREY M. and DALE L. 

HUTCHINSON. 1987 Interim Report 
on Archaeological Research at the Tat- 
ham Mound, Citrus County, Florida: 
Season III. Florida Museum of Natural 
History, Miscellaneous Project Report 
Series, No. 30, Gainesville. 

MORGAN, GARY S. 1991. Neotropical 
Chiroptera from the Pliocene and 
Pleistocene of Florida. Bulletin of the 
American Museum of Natural History 
206:176-213. 

NABHAN, GARY. 1987. The origins of 
neotropical horticulture following 
megafaunal extinctions: Did humans 
disperse and select anachronistic 
fruits? Abstracts, 10th Annual Meet- 
ing of the Society of Ethnobiology, 
Gainesville, Florida. 

NEE, MICHAEL. 1990. The domestication 
of Cucurbit a (Cucurbitaceae). in New 
perspectives on the Origin and Evo- 
lution of New World Domesticated 
Plants. Peter K. Bretting (editor). Eco- 
nomic Botany 44(3 SUPPLEMENT): 
56-68. 



NEWSOM, LEE A. 1987. Analysis of 
botanical remains from Hontoon 
Island (8Vo202), Florida: 1980-1985 
excavations. The Florida Anthro- 
pologist 40:47-84. 

.. 1991. Plant remains from the 

Ribera Gardens site, St. Augus- 
tine, Florida, in Archaeology at Ribera 
Gardens. Stanley Bond (editor). St. 
Augustine Preservation Board, St. 
Augustine. Manuscript on file, Florida 
Museum of Natural History. 

1993. Horr's Island archae- 

obotanical research. Pp. 591-643 in 
Final Report on Horr's Island: The 
Archaeology of Archaic and Glades 
Settlement and Subsistence Patterns. 
Michael Russo (editor). Institute of 
Archaeology and Paleoenvironmental 
Studies, Monograph 3, Gainesville, 
Florida, in press. 
and DEENA S. DECKER. 



1986. Archaeological Cucurbitaceae in 
peninsular Florida. Paper presented at 
the 43rd Annual Meeting of the South- 
eastern Archaeological Conference, 
Nashville, Tennessee. 



NEWSOM 



PURDY. 1990. Florida canoes. Florida 
Anthropologist 43:164-179. 

NEWSOM, LEE A. and IRVY R. QUIT- 
MYER. 1992. Archaeobotanical and 
faunal remains from Fig Springs Mis- 
sion (8Col). Pp. 206-243 in Excavation 
on the Franciscan Frontier: Archaeol- 
ogy at the Fig Springs Mission. Brent 
R. Weisman (editor). University 
Presses of Florida, Gainesville. 

PETERSEN, JAMES B. 1991. Archaeolog- 
ical Testing at the Sharrow Site: A 
Deeply Stratified Early to Late Holo- 
cene Cultural Sequence in Central 
Maine. Occasional Publications in 
Maine Archaeology, No. 8, Maine Ar- 
chaeological Society and the Maine 
Historic Preservation Commission, 
Augusta. 

PURDY BARBARA A. 



1987. Investiga- 
tions at Hontoon Island (8Vo202), 
an archaeological wetsite in Volusia 
County, Florida: An overview and 
chronology. The Florida Anthropol- 
ogist 40:4-12. 
_. 1988. Archaeological wet 

sites: Untapped archives of prehis- 
toric documents. Pp. 325-335 in Wet 
Site Archaeology. Barbara A. Purdy 



96 



NEWSOM, WEBB 



Vol. 13, No. 1 



(editor). Telford Press, Caldwell, New 
Jersey. 

1991. The Art and Archae- 
ology of Florida's Wetlands. CRC Press, 
Boca Raton, Florida. 

RUSSO, MICHAEL, BARBARA A. 
PURDY, LEE A. NEWSOM, and RAY 
M. MCGEE. 1992. A reinterpretation 
of late Archaic adaptations in central- 
east Florida: Groves' Orange Midden 
(8Vo2601). Southeastern Archaeology 
11:95-108. 

SCARRY, C. MARGARET. 1985. The use 
of plant foods in sixteenth-century St. 
Augustine. The Florida Anthropolo- 
gist 38:70-80. 

1991a. Plant remains from 

Fort Matanzas. Report on file, Depart- 
ment of Anthropology, Florida State 



University. 

1991b. Plant production and 

procurement in Apalachee Province. 

The Florida Anthropologist 44(2-4): 
285-294. 

. and LEE A. NEWSOM. 

1992. Archaeobotanical research in the 
Calusa heartland. Pp. 375-401 in Cul- 
ture and Environment in the Domain 
of the Calusa. William H. Marquardt 
(editor). Institute of Archaeology and 
Paleoenvironmental Studies, Mono- 
graph 1, Gainesville, Florida. 

SMITH, BRUCE D. 1987. The independent 
domestication of indigenous seed 
bearing plants in eastern North Amer- 
ica. Pp. 3-47 in Emergent Horticultural 
Economies of the Eastern Woodlands. 
William F. Keegan (editor). Center for 
Archaeological Investigations, Occa- 
sional Paper, No. 7, Southern Illinois 
University, Carbondale. 

1992. Rivers of Change. 

Institution Press, Wash- 



ington, DC. 



, C. WESLEY COWAN 



MICHAEL P. HOFFMAN. 1992. Is it 
an indigene or a foreigner? Pp. 67-100. 
in Bruce D. Smith: Rivers of Change. 
Smithsonian Institution Press, Wash- 
ington, DC. 

TOMLINSON, P.B. 1980. The Biology of 
Trees Native to Tropical Florida. Har- 
vard University Printing Office, 
Allston, Massachusetts. 

VERNON, R.O. 1951. Geology of Citrus 
and Levy Counties, Florida. Florida 



Geological Survey Bulletin, No. 33, 
Tallahassee. 

WALKER, KAREN J. and WILLIAM H. 
MARQUARDT (editors). 1994. Pine- 
land Archaeology. Institute of Archae- 
ology and Paleoenvironmental Studies, 
Monograph 4, Gainesville, Florida, in 
preparation. 

WALTERS, TERRENCE W. and DEENA S. 
DECKER-WALTERS. 1993. Systemat- 
ics of the endangered Okeechobee 
gourd (Cucurbita okeechobeensis: Cucur- 
bitaceae). Systematic Botany 18:175- 

187 

WATSON, PATTY JO. 1989. Early plant 
cultivation in the eastern woodlands 
of North America. Pp. 555-571 in For- 
aging and Farming: The Evolution of 
Plant Exploitation. David R. Harris 
and Gordon C. Hillman (editors). 
Unwin Hvman, London. 

WATTS, WILLIAM A. and BARBARA 
C.S. HANSEN. 1988. Environments 
of Florida in the late Wisconsin and 
Holocene. Pp. 307-323 in Wet Site 
Archaeology. Barbara A. Purdy (edi- 
tor). The Telford Press, Caldwell, New 

Jersey. 

and E.C. GRIMM. 1992. 

Camel Lake: A 40,000-year record ot 
vegetational and forest history from 
Northwest Florida. Ecology 73:1056- 

1066. 
WEBB. S. D, 



sion in 



North American terrestrial 
mammal communities. Pp. 181-203 in 
Patterns in the Structure of Mam- 



malian Communities. 



W. Morris, 



Z. Abramsky, B.J. Fox, and M.R. Wilhg 
(editors). Special Publications of the 
Museum, No. 28, Texas Tech Univer- 
sity, Lubbock. 

_J _. 1991. Ecogeography and the 

Great American Interchange. Paleo- 
biology 17:266-280. 
JERALD T. MILANICH, 

ROGER ALEXON, and JAMES 5. 
DUNBAR. 1984. A Bison antiquus kill 
site, Wacissa River, Taylor County, Flor- 
ida. American Antiquity 49:384-392. 



WEBB 



1991. Diet of Manumit atncricamtni in 
late Pleistocene of Florida. Journal o 
Vertebrate Paleontology Abstracts H 
(3 SUPPLEMENT):60. 
WEBB, S. DAVID and K.T WILKINS. 1984. 
Historical biogeography of Florida 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



97 



Pleistocene mammals. Special Pub- 
lication of the Carnegie Museum of 
Natural History, No. 8:370-383. 
WHITAKER, THOMAS W. and G.F. CAR- 
TER. 1946. Critical notes on the origin 
and domestication of the cultivated 
species of Cucurbit a. American Journal 



of Botany 33:10-15. 



WHITAKER, THOMAS W. and HUG! I C. 
CUTLER. 1971. Prehistoric cucurbits 
from the Valley of Oaxaca. Economic 
Botany 25:123-127. 

and RICHARD M ACNEISI 1 

1957 Cucurbit materials from the 
caves near Ocampo, Tkmaulipas. 
American Antiquity 22:352-358. 



BOOK REVIEW 



Household Ecology: Economic Change and Domestic Life Among the Kekchi 
Maya in Belize. Richard R. Wilk. Tucson and London: University of Arizona 
Press, 1991. $55.00 (clothbound). Pp. xx,280. ISBN 0-8165-1214-0. 



more 



readable book (although "household ecology" is arguably redundant). This is a 
comprehensive treatment with ten informative figures and 28 tables of data— a 
reflection of Wilk's familiarity with the study area, the Kekchi people and their 
interactions with their natural and agricultural ecosystems, themselves and with 
the so-called "outside world. 



r? 



The extended preface is informative and it should not be skipped. In the intro- 



Wilk 



by an outline of the book, chapter by chapter: Household social evolution; The 

- . «« ._• T-»* 1 



The 



ical setting; Land tenure and crops; Domestic animals; Hunting and gathering; 



nomic 



ecology. 



Predictably, this book will hold greater appeal for professional anthropologists 
than ethnobiologists, and I feel certain the author had the former group in mind as 
the primary targeted readership. Nonetheless, as a general ethnobotanist, I found 
the last half of the book, and especially chapters 6 through 9, meaty with concepts 
and discussions not only interesting but also germane to studies in ethnobiology 
and I suggest that investigators in this discipline might find ideas to consider in 

their research. 

Although the price of the book will perhaps discourage at least some ethno- 



from 



important acquisition to institutional libraries. 



Willard Van Asdall, Past E< 
Journal of Ethnobiology 

4479 N. Summer Set Loop 
Tucson, AZ 85715, USA 



BOOK REVIEW 



Traditional Plant Foods of Canadian Indigenous Peoples: Nutrition Botany 
and Use. Harriet V. Kuhnlein and Nancy J. Turner. Volume 8 in the Food and 



98 



BOOK REVIEW Vol. 13, No. 1 



Solomon 



York and Philadelphia: Gordon and Breach Science Publishers, 1991. Pp. xi, 



members 



maps. $88.00 (hardbound), $38 



Ethnobotanical studies amongst Aboriginal Peoples in most parts of North 
America are "few and far between." It is indeed encouraging that such a highly 
detailed and thorough report on food plants and nutrition of the Indigenous Peo- 
ples of Canada has appeared. The two authors— a well known nutritionist from 
McGill University and a widely respected ethnobotanist— are to be congratulated 
for their presentation of such a masterly contribution. 

The forward by Laurie Montour, a Native Canadian and a staff member of the 
Assembly of First Nations, sets the tone of the book: "We need to work hard 
together to preserve our knowledge and to protect the environments of the plant 
foods of the world's indigenous people. This book is a good step along the way." In 
the Acknowledgments, the authors indicate the great number of individuals 
aboriginal consultants, students, governmental agencies, and scientific colleagues 
whose contributions have helped make this book the encyclopedic masterpiece 
that it is. The Introduction, Chapter 1, tells us that "The scientific literature was 
searched for nutrient information of approximately 1,050 species that were identi- 
fied as edible and available in Canada," and that "... nutritional, botanical and 
ethnological data for more than 1,000 species of edible plants" are included. 

There follows Chapter 2, "What's So Special About Indigenous Foods?" 
Amongst other "specialties" is the usefulness of information in "genetic research, 
in enhancing existing crops or . . . developing new ones." Chapter 3 deals in 
depth with "An Overview of the Nutrient Value and Use of Plant Foods by Indige- 
nous Peoples." In the fourth chapter— a major section of the book— are considered 
the botany and methods of use of indigenous plant foods of Canada. A compre- 
hensive list of plant food species makes up Chapter 5, a convenient tabular sum- 
mary of earlier chapters. Another tabular chapter deals with "Nutrient Values of 
Traditional Plant Foods"— a most useful addition that occupies 162 pages. 

The Bibliography comprises 519 items and is followed by three Appendices: 
Linguistic Affiliations and Locations of Indigenous Peoples of Canada (with three 
maps); Species by Common Name; and Species by Botanical Name. The Index 
contains both common and scientific plant names, and nutritional and other 
chemical constituents of the food plants. 

This volume will certainly long remain an example of the very finest in eth- 
nobotanical literature. Furthermore, its utility will be evident as a manual for 
quick consultation by specialists, as a valuable text or classroom reference work, 
and as an instrument furthering the development of the interdisciplinary field of 



ethnobotany. 



Richard Evans Schultes 

Director Emeritus 

Botanical Museum of Harvard University 

Cambridge, Massachusetts 



/. Ethnobiol. 13(1):99-130 



Summer 1993 



NEWS and COMMENTS 



Notes may be submitted on CONFERENCES AND SEMINARS 
REQUESTS FOR INFORMATION in addition to the sections included b 
Because the Journal is published only twice a year, dated items must be receiv 
least six months in advance of the event. 



BOOK REVIEWERS NEEDED 



The following titles have been received for review in the Journal of Ethnobiobgy. 



Human Adaptation in the Upper Mississippi Valley: A Study of the Pamm 
Creek Oneota Site (47Lc61) La Crosse, Wisconsin. Constance M. Arzigia 
Robert F. Boszhardt, James L. Theler, Roland L. Rodell, and Michael J. Scott. T 
Wisconsin Archaeoloeist. Volume 70 (Nos. 1-2, 1989). Monograph of the M 



Wisconsin-La Crosse [W: 
Box 1292. Milwaukee, Wi 



$12 



Tree of Life: Biology, Utilization and Conservation. Michael J. 
Balick (Editor). Proceedings of the Symposium at the 1986 Annual Meeting 
of the Societv for Economic Botanv held at the New York Botanical Garden, Bronx, 



The 



$53 



M. Boom. Bronx, NY: The 



68. $15 



Herbal Dentistry. Herbal Dental Remedies from Ancient Times to the Present 
Day. Joseph G. Carter and William J. Carter. Chapel Hill: The University of North 
Carolina, 1990. Pp. 77 (typescript). $17.95 (softcover). ISBN 0-930989-01-8. 

Folk Dentistry. Cultural Evolution of Folk Remedies for Toothache. Joseph G. 
Carter and William J. Carter. Chapel Hill: The University of North Carolina, 1990. 



$17.95 



Marana Community in the Hohokam World 

John H. Marden 



anajonn n. Maraen (taitors;. /\ntnrupuiugn-m i apcic, »-»«.«~- — 

University of Arizona Press, 1992. Pp. 121. Price ? (softcover). ISBN 0-8165-1314-7. 

The Ethnobotany of Southern Balochistan, Pakistan, with Particular Reference 



Medicinal Plants. Steven M 



84 



M 



W 



100 



NEWS and COMMENTS Vol. 13, No. 1 



ington & Indianapolis: Indiana University Press, 1991. Pp. 280. $45.00 (paper- 
bound). ISBN 0-243-31979-X. 

Agriculture and the Onset of Political Inequality before the Inka. Christine 
A. Hastorf. New York and Cambridge: Cambridge University Press, 1993. 
Pp. xv,298. $69.93 (hardback). ISBN 0-521-40272-7. 

The Tasaday Controversy: Assessing the Evidence. Thomas N. Headland (Editor). 
Washington, D.C.: American Anthropological Association Special Publication 
No. 28. 1992. Pp. xi,255. $19.95 (paperbound). ISBN 0-913167-51-7 

The State of Nature: Ecology, Community, and American Social Thought, 
1900-1950. Gregory Mitman. The University of Chicago Press, Chicago, Illinois, 
1992. Pp. 290. Price ? (clothbound). ISBN 0-226-53236-4. 



Butterflies of the Bulolo-wau Valley. Michael Parsons. Handbook No. 12 of 
the Wau Ecology Institute, Wau, Papua New Guinea. Honolulu: Bishop Museum 
Press, 1991. Pp. 280. $30.00 (paperbound). ISBN 0-930897-61-7. 

New Directions in the Study of Plants and People: Research Contributions 
from the Institute of Economic Botany. Ghillean T. Prance and Michael J. Balick 
(Editors). Bronx: The New York Botanical Garden, Scientific Publications Depart- 
ment, 1990. Pp. 278. $55.00 (paperbound). ISSN 0741-8280. 

Phytolith Systematics: Emerging Issues. George Rapp, Jr., and Susan C. 
Mulholland (Editors). New York: Plenum Press, 1992. Pp 350. $49.50 (hardcover). 
ISBN 0-036-44208-6. 

Rivers of Change: Essays on Early Agriculture in Eastern North America. 



Washin 



5 



M 



W. Stoffle, David B. Halmo, John E. Olmsted and Michael 



S 



Mich 



World 



van Zeist, Krystyna Wasylikowa 



Work Group for Palaeoethnobotany. Willem 
nd Karl-Frn<;t RphrP (Editors). Rotterdam, 



Vermont: A. A. Belkema, 1991. Pp. ix,350. $60.00 
(hardcover). ISBN 90-6191-881-2. 

Persephone's Quest: EntheOgens and the Origin of Religion. R. Gordon 
Wasson, Stella Kramrisch, Jonathan Ott, and Carl A. P. Ruck. New Haven, Con- 
necticut: Yale University Press, 1986. Pp. 257. $37.00 (clothbound). ISBN 



0-300-03877-1. 



If you would like to review any of these books an 



review completed within four months after receiving the book, please write to: 



your 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 101 



Nancy J. Turner 
Book Review Editor 

Journal of Ethnobiology 

Environmental Studies Program 

P.O. Box 1700 

University of Victoria 

Victoria, British Columbia 

CANADA V8W 2Y2 

Phone:(604)721-6124 



PROJECTS AND PROGRAMS 



World Wide 



[WWF] 



Organization [UNESCO] and the Royal Botanic Gardens, Kew. Through the ini- 
tiative, these organizations seek to promote the sustainable and equitable use of 
plant resources by providing support to ethnobotanists from developing coun- 
tries. People and Plants field personnel organize participatory workshops, make 
advisory visits, and provide literature on ethnobotany, traditional ecological knowl- 
edge, and sustainable plant resource use. Field activities build on work within 



reserves 



projects. For more information contact any one of the three partners: (1) Biodiver- 
sity Unit, Conservation Policy Division, WWF International, World Conserva- 
tion Centre, Avenue du Mont-Blanc, 1196 Gland, SWITZERLAND; FAX 
41.22.364-8219; (2) Division of Ecological Sciences, Man and the Biosphere Pro- 
gramme, UNESCO, 7 place de Fontenoy, 75732 Paris CEDEX 07 SP, FRANCE; Fax 
33.1.40659897; (3) The Director, Royal Botanic Gardens, Kew, Richmond, Surrey 
TW9 3AB. UNITED KINGDOM; Fax 44.81.332-5197. 



JOURNALS AND OTHER MEDIA 



Conservation 



Smithsonian Institution, contains 4-6 pages of short contributed articles; notes on 
educational materials, new publications, and courses; announcements of meet- 
ings and job opportunities; and citations of current literature. It is currently being 

' _ rr , . r-r, i.-^ tv* *.«,^i^o tho month Iv npws- 



more 



1.202.357-2027; Fax 1.202.786-2563. 
Volume 1 /Number 1 (February 
velopment Monitor (IK&DM) has j 
the IK&DM is a publication of and fc 



Conservation N 
3 166, Washingt 



IK&DM 



tuc llllCItbltU III IIlUIV:fIlUU3 Miuvvivw & v.. . - 

for International Research and Advisory Networks (CIRAN) ,n close coopera ,on 
with the Centre for Indigenous Knowledge for Agriculture and Rural Develop- 

°. - , j ru„Jnnmpnf Program LhAU . 



ment 



and national and regional Indigenous Knowledge Resource Centres. The IK&DM, 



102 



NEWS 



Vol. 13, No. 1 



which replaces CIKARD News, is published three times a year, in two regular 
issues and one special issue. Issue no. 1, a forty-paged double issue, has already 
been sent to 3,000 people in 130 countries. Each issue contains articles describing 
the application of indigenous knowledge in development projects in many parts 
of the world. A section called Communications includes information on resource 
centers, networks, research, conferences, databases, and publications. For more 
information, contact: CIRAN, c/o Nuffic P.O. Box 90734, 2509 LS The Hague, The 
Netherlands. Telephone 31.70.351-0574; Fax 31.70.351-0513. 

The Forest, Trees and People Newsletter is a quarterly publication, available 
in French, English, and Spanish, that provides a forum to discuss social, eco- 
nomic, and political issues of relevance to community forestry. This newsletter is 
part of the networking actions of the FTP Programme, a multi-donor, FAO-coordi- 
nated initiative which supports a wide variety of activities from policy formulation 
to field projects through regional initiatives in Asia, Africa, and Latin Al ^ eric ^ 
The experiences of the programme have been documented in 
lications that are distributed through the network. 

Each issue of the newsletter, spanning some 50 to 60 pages, contains numer- 
ous articles describing case studies and approaches which can be used to support 
local farmers' own initiatives to manage their natural resources in a sustainable 
manner. There are separate sections on information, activities, and events; re- 
views of literature; editorials and letters. For more information, write: Editor, Fit 
Newsletter, IRDC, Swedish University of Agricultural Sciences (SUAS), Box 7005, 
S-75007 Uppsala, Sweden. 



more 



COURSES AND DEGREES IN ETHNOBIOLOGY 



A request was made in the last issue of the Journal for current information on 
university-level degree programs and courses in ethnobiology. If you know ot a y 
such programs or courses, whether in the United States or in a foreign country, 
please contact the editor of News and Comments. 



COMMENT 



USE OF POLLEN CONCENTRATIONS IN COPROLITE ANALYSIS: 



VIEWPOINT WITH A COMMENT 



>/< 



rt- 



)f Biology, University of New Mexico, Albuquerque, NM 



INTRODUCTION 

To judge from the phrasing of their article's title, "Use of pollen concen ?^j 
in paleopharmacology: Coprolite evidence of medicinal plants," Reinhard e ^ 
(1991) proposed to demonstrate prehistoric medicinal plant use in their anal y slS tu 
pollen spectra from human coprolites by means of pollen concentrations. In a 
ality, the pollen data were presented in two forms: pollen concentrations 
relative frequencies (relative percentages). 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 103 



Two brief but technical discussions will explore the implications of the choice 



me 



coprolite analysis, before undertaking a re-examination of selected pollen data 



some 



POLLEN ANALYTICAL TECHNIQUES 



Relative frequencies were used by Reinhard et al. (1991) to display and ana- 
lyze their coprolite pollen data, despite the title of their article. This technique is 
based on the work of Barkley (1934), Dimbleby (1957), Martin (1963), von Post 
(1918, 1967), and others. Relative frequencies are obtained by counting the pollen 
grains in a preparation to a total of 200 (or more), and then dividing the number of 
grains of each pollen type seen by the total and multiplying by 100. Converting 
pollen data to relative frequencies has two major effects. One, the conversion of 
raw data to a standard percentage masks variations of total pollen abundance in 



com 



in relative frequency in response to a decrease or increase in any other taxon in 
that sample (Birks and Gordon 1985:11). 

Pollen concentrations were cited by the authors only for the total concentra- 
tion of each sample; these data were not used in any of the analyses they reported. 
Although not referenced in the article, the concept of determining the concentra- 
tion of pollen grains/unit of sample is based primarily on the work of Benninghof 
(1962) and Maher (1981). The method depends on the addition of known numbers 
of exotic marker grains (commonly called "spike grains") to samples prior to labo- 
ratory extractionfboth pollen grains and spike grains are tallied separately un 



Estimatio 



sam 



# pollen grains/unit sample = # fossil pollen counted x # of spike grains added 

# spike grains counted weight (or vol) of sample 

The method can produce large numbers if pollen in the 8 ^^^ 
The unique strength of the pollen concentration approach ««?^^CTa« 
.nml^f Aitf-rS ,™„nt<Tnf nollen/unit sample, especially for individual taxa 



in the Dollen snectrum 



round to the same values as when raw pollen counts are divioea py i J f 

count to produce relative frequencies, thus losing the parfcular advantages 

scale provided by pollen concentrations. , among 

Both relative frequencies and pollen concentrates ar m « ^ *", J 
penologists to express pollen data. Data from both W™ * es « h P this 

for fho c.™ M f „f ct™1oq as done bv Reinhard et al. (iwij, cm 5 r 



same set ot sam 
tice may be more common 
samples. 



^nerally speaking, three factors influence **g»tttt 
from a sample: rate of pollen production of the contributing vege^ ^ 

pollen (and sediment) accumulation at the sampled locus rf 



at the sam 



104 



NEWS and COMMENTS Vol. 13, No. 1 



these events, pollen production and accumulation, are crucial to the present dis- 



cussion. 



Environmental palynologists have used relative frequencies to study dia- 
chronic vegetation trends in samples from bogs and other open-air sites since the 
earliest days of palynology, under the fundamental assumption that the pollen 
rain falling on a given locality is basically uniform from year to year (Birks and 
Gordon 1985:3-4). Relative frequencies smooth over minor variations in a larger 
environmental pattern, and are the traditional method of choice for analyzing 
environmental samples under this theoretical construct. Critical to the success of 
this approach, however, is the taking of the environmental samples from areas 
undisturbed by human activities, especially archaeological sites. This is because 
environmental palynologists recognize that human activities, past and present, 
disturb the record of environmental pollen rain. Disturbance can take the form of 
physical turbation of sediments, for example, which can often be seen in the wall 
of a trench dug for the purpose of sampling. Other past human disturbance may 
be invisible, such as an alteration of the localized pollen rain as the result of agri- 
cultural activities. 

Archaeological palynologists, in contrast, take samples to answer research 
questions about prehistoric human behavior. Archaeological palynology strives to 
pinpoint past human interference in an otherwise natural pollen rain: differences 
in rate of pollen accumulation are one clue to such human interference; another is 
presence of pollen grains of cultivated plants. The best archaeological samples for 
studying prehistoric human behavior come from enclosed sampling loci such as 
floors, pits, vessels, and the like. These are sampling loci specifically excluded 
from environmental pollen rain (commonly viewed as "contamination" or "noise") 
falling on the open ground. Total numbers of pollen grains in such samples often 
vary wildly from sample to sample, because human behavior is the primary source 
of the pollen selectively captured by these sampling loci. 

For example, a pit was used to store pollen-dusted seeds or plant parts, in 
effect accelerating the rate of pollen accumulation there. An adjacent pit was used 
to store other items, in effect retarding the rate of pollen accumulation there. Both 
events took place within a pit structure, the roof of which protected the interior 
from the natural pollen rain of the site area. Samples scraped from the bottom of 
both pits will yield pollen spectra skewed not only in terms of total pollen abun- 
dance, but also in favor of individual pollen taxa. This is the beginning of a pictur" 
of prehistoric human behavior in connection with the two pits, but the choice of 
data presentation, relative frequencies or pollen concentrations, will determine 
what patterns are actually seen by the analyst. 

Pollen concentrations will highlight differences in pollen accumulation be- 
tween the samples in this example. These differences in pollen accumulation w* 1 
be lost if data are expressed as relative frequencies. Perhaps more importantly 
rare pollen taxa can be analyzed as real numbers/unit of sample when expressed 
as pollen concentrations because the data need not sum to a predetermined total 
such as 100%. Rare pollen taxa can only reflect 1% or less of a 200-grain count 
when expressed as relative frequencies. It is this lack of ability to highlight rare 
pollen taxa that is the greatest shortfall of relative percentages as applied to ar- 
chaeological samples taken to answer human behavioral questions. 



e 



ill 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 1(K 



POLLEN IN HUMAN COPROLITES 



Pollen grains enter the body by inhalation from the air and by ingestion via 

beverages and food. The rate at which pollen grains and other items exit the 

experimental data (Alvarez and I reed- 



stem is another matter. Modern 
14: Kelso 1976: Martin 1965: Willi 



m 



j f _ _ : _ _ : _. 

from day to day. In order for coprolite analysis to proceed at all, a number of basic 
assumptions are made; it is recognized that they are a gross oversimplification of 
demonstrated variability. 

It is assumed that the pollen spectrum passing through a person's system 
varies from day to day. The recovery of large amounts of pollen from more than 



im 



one day (assuming that all the coprolites under study were deposited over time, 
and not as one single event). The natural windborne pollen rain should cause a 
relatively uniform concentration of "background" pollen types, because they are 
seasonally present in the air for more than one day. The ingestion of windborne 
and insectborne pollen types with food and drink, on the other hand, should 
produce a pollen assemblage independent of the windborne pollen rain. 



more 



resulting from "accidental" or "incidental" ingestion as opposed to "purposeful" or 
"economic" ingestion because of limitations in the English language. It is assumed 
that pollen was probably rarely ingested as a recognized substance per se. The 
distinction to be made is whether pollen was inhaled or ingested in pollen-dusted 
water or other liquid (incidental ingestion), or whether pollen accompanied the 
deliberate ingestion of plant material that happened to be pollen-bearing (economic 
or purposeful ingestion). Pollen resulting from so-called purposeful ingestion is 
viewed as indicating dietary, medicinal, or other behavior involving the parent 
plant. Unusual pollen abundance is one clue to an economic use of plants; presence 
of insectborne pollen types is another, because these types are rare or absent in the 
atmospheric pollen rain. In order to establish some sort of baseline for comparison, 



samples from the air or the modern 



centration 



sample); these figures provide an idea of how many pollen grains are available to be 
accidentally inhaled by local residents today (and in the past, by extension). 

It is apparent that human coprolites are very far removed from the traditional 
environmental concept of a "uniform pollen rain," and constitute instead an ex- 
treme class of archaeological pollen samples influenced by idiosyncratic human 



means 



studied, then a technique which focuses on different pollen accumulation will 



give better data. 



HUMAN 



Reinhard et al. (1991) chose to analyze the pollen present in coprolites, the 
ultimate behavioral samples, using methods designed for environmental pal- 
ynology and its assumptions (for the record, my own work with prehistoric and 



106 



NEWS and COMMENTS Vol. 13, No. 1 



TABLE 1.— Approximate pollen concentrations, in number of pollen grains/g of 
sample, of Bighorn Cave coprolites. 



Concentration 1 Lab No. Ephedra % Salix % Larrea % 



12 

14,300 1 429 3 572 4 143 1 

16,100 2 322 2 322 2 

17,100 17 171 1 1,710 10 171 1 

20,500 ■ 14 410 2 

26,000 6 



- 820 4 

520 2 780 3 

26,200 5 524 2 524 2 262 1 

29,300 9 879 3 4,395 15 

29,400 3 - - 1,176 4 588 2 

36,300 10 726 2 363 1 1,452 4 

53,000 13 530 1 530 1 2,650 5 

72,300 4 4,338 6 1,446 2 

114,900 7 

129,000 18 

150,000 8 



2,580 2 

1,500 1 

224,000 11 • — — 192,640 86 

1,137,000 16 — - 693,570 61 



means of 21 samples 



4,338 


6 


3,870 

i 


3 


1,680,000 


75 


1,895,400 


81 


170,838 


8.6 



2,240,000 20 

2,340,000 21 

4,973,000 15 - - 4,973,000 100 

5,000,000 19 



22,400 1 67,200 3 
_ _ 46,800 2 



280,821 14.0 5,756 1-2 



'"Total concentration" (column 1) and relative frequencies taken from Reinhard et al. (1991:123). 
Approximate concentration values for each taxon calculated from relative frequencies as follows: total 
concentration x relative frequency of a taxon = concentration for that taxon 



modern coprolites also used relative frequencies to express the polle 
[Williams-Dean 1978]; this work is slated for revision). It is my position thai 



meant 



missed 



I converted the relative frequency data presented by Reinhard et al. (1991) to 



m 



for each pollen taxon, yielding a close approximation of the actual pollen con- 
centrations (Table 1, 2). Original relative frequency values are included for sake of 



mask 



many 



convert- 



ing the data for each pollen taxon to pollen concentration provides a much richer 
picture of prehistoric plant use in each sample. 

Reinhard et al. (1991) chose to spotlight three genera of plants with known 
ethnopharmacological uses: Larrea (creosote bush), Salix (willow), and Ephedra 
(Morman tea). Data on other pollen types recovered from the coprolite specimens 
were largely omitted. Only the pollen data from Biehorn Cave. Arizona, and the 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



107 



TABLE 2.— Approximate pollen concentrations, in number of pollen grains/g of 
sample, of NAN Ranch burial coprolite and burial soil samples. 



Taxon 



Burial 
Coprolite 



% 



Burial 
Soil 1 



% 



Burial 
Soil 2 



% 



Burial 
Soil 3 



o o o 



°/c 



Salix 

Apiaceae 

Artemisia 

Brassicaceae 

Cactaceae 

Cucurbita 

Cheno/Am 

Fabaceae 

high-spine aster 

low-spine aster 

Pinus 

Poaceae 
Typha 

unidentifiable 
Zea 



117,000 
2,250 



26 
tr 



93 
93 



tr 

tr 



*• 



238,500 


53 


4,500 
36,000 


1 
8 


2,250 


tr 



63 



tr* 



186 



1 



63 



tr 



13,020 



70 



7,056 



56 



4,500 

4,500 

2,250 

45,000 



1 
1 
tr 

10 



744 
2,046 



4 
11 



63 
63 
63 



3,886 
34 



58 
tr 



2,142 



tr 

tr 

tr 

17 



67 
201 
804 



1 

3 

12 



744 
1,674 



4 

9 



63 
2,646 



tr 

21 



** * 



469 
1,206 



7 
18 



TOTALS: 
Pollen counted 
Concentration 



1,011 
450,000 



209 
18,600 



211 

12,600 



124 
6,700 



• * 



*** 



o o 



o o o 



tr trace (here assumed to be a uniform 0.5%). 
trace % reported for Burial Soil 1 by Shafer et al. (1989:22; Sample 2). 
trace % reported for Burial Soil 2 by Shafer et al. (1989:22; Sample 3). 

trace % reported as 5% in Shafer et al. (1989:22); concentration value would equal 630 grains/g. 
provenience given as "grave fill" (Shafer et al. 1989:22; Sample 2). 
provenience given as "soil from atop left pelvis" (Shafer et al. 1989:22; Sample 3). 
provenience given as "soil from beneath pelvis" (Shafer et al. 1989:22; Sample 4). 

Values do not necessarily sum to total. Approximate concentration values calculated as in Table 1. 
"Total pollen grains counted" taken from Shafer et al. (1989:22). "Total pollen concentration" taken 
from Reinhard et al. (1991:127). 

NAN Ranch, New Mexico, are addressed here, since the Texas samples discussed 
by Reinhard et al. (1991) did not have pollen concentration values available for 
their pollen spectra. The conclusions from Reinhard et al. (1991) and my reanaly- 
sis are summarized and compared in Table 3. 

Bighorn Cave data. -The authors stated as an underlying assumption that insect- 
borne pollen types were to be viewed as signalling "... the intentional con- 
sumption of dietary or medicinal plants" (page 122) only when present in relative 
frequencies greater than 4%. To further highlight intentional consumption of pol- 
len grains, the authors calculated the mean occurrence of the three pollen taxa 
under discussion (Larrea, Salix, and Ephedra) for all 21 Bighorn Cave coprolites; 
these means were calculated on the basis of relative frequencies, not pollen con- 
centration values. 



108 



NEWS 



Vol. 13, No. 1 



TABLE 3.— Comparison of conclusions: relative frequencies vs. pollen 



concentrations 



Bighorn Cave Relative Frequencies 

(1) Larrea pollen was accidentally ingested 
in the 10 specimens in which it was seen 
(mean of relative frequencies was below 
4%). 

(2) Salix pollen frequencies were signifi- 
cant in only three of 16 samples contain- 
ing Salix (relative frequencies were above 
the mean of 14%). 



(3) Relative frequencies of Ephedra in sam- 
ples from the modern ground surface 
were 8%-45%, suggesting that wind- 
borne Ephedra pollen was accidentally 
ingested or inhaled in 10 of 12 samples 
containing Ephedra. 



NAN Ranch Relative Frequencies 

(1) Control samples from the midden con- 
tained more pollen from Cheno/ Ams, 
Poaceae, and low-spine Asteraceae than 
did the coprolite, indicating that the lower 
amounts of these pollen types in the 
coprolite were derived from the natural 
pollen rain in the site area. 

(2) Midden samples contained more pol- 
len from corn than did the coprolite, 
pointing up the unusual nature of the 
last meal(s) of the individual. 

(3) The pollen spectra of the three midden 
samples resembled each other more than 
the coprolite pollen spectrum; these were 



Bighorn Cave Pollen Concentrations 

(1) Larrea pollen was not accidentally 
ingested (relative frequencies represent 
from 143 to over 67,000 insect-transported 
Larrea pollen grains/g of sample). 

(2) Salix pollen concentrations were sig- 
nificant (above an arbitrary 1000 grains/g) 
in 10 of 16 samples containing Salix. Ex- 
perimental data are needed to suggest the 
number of Salix pollen grains available for 
accidental ingestion. 

(3) Concentration values of Ephedra in two 
New Mexico samples from the modern 
ground surface suggest that windborne 
Ephedra pollen might have been acciden- 
tally ingested or inhaled in only 8 of 12 
samples containing Ephedra. Better data are 
needed from modern plant communities. 

NAN Ranch Pollen Concentrations 

(1) The coprolite sample contained more 
pollen from Cheno/ Ams, Poaceae, and 
low-spine Asteraceae than did the sam- 
ples from the midden, despite the mid- 
den's exposure to the atmospheric pollen 
rain, indicating that the coprolite pollen 
spectrum is behavioral in origin. 

(2) The coprolite sample contained more 
pollen from corn than did the midden 
samples; no unusual last meal(s) are indi- 
cated in the coprolite pollen spectrum. 

(3) Burial Soil Sample 2, taken from atop 
the left pelvis, contained up to twice as 
many grains of Zea pollen as the other two 



lumped and interpreted as environmental midden samples, suggesting that the sam 



or non-behavioral control samples. High 
relative frequencies of Zea pollen in sam- 
ples 2 and 3 were suggested as reflecting 
ritual preparation of the grave with corn 



corn 



(1989:27), but not by Reinhard et al. (1991). 



pie reflected decomposed intestinal con- 
tents. The concentrations of Zea pollen 
in the other two midden samples reflect 
localized deposition of pollen-bearing 
plant material which may predate the 
grave, or may have resulted from ritual 
use of corn products in the grave as sug- 
gested by Shafer et al. (1989:27). 



— H 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 109 



mean 



and all Larrea data from the coprolites were dismissed by Reinhard et al. (1991) as 
". . . accidentally ingested" (page 125). However, when pollen concentration values 



mean 



21 (67,200 and 46,800 grains/g, respectively) depart significantly from the mean 



more 



ble 1). In fact, most of the calculated concentration values for insectborne Larrea 
pollen grains appear to be high enough to indicate intentional ingestion regard- 



from the mean 



mo 



// 



environmental" or "nat- 



ural" baseline from which to infer "behavioral" departures. 

My calculation of concentration values and the mean for Salix pollen data from 
Bighorn Cave coprolites indicates that only two samples (samples 15 and 16), not 



sam 



The 



Sampl 



attention to it if I were conducting this analysis de novo. In fact, given the sampling 



most 



Salix pollen appear to be high enough to indicate possible intentional ingestion. 
At this point, I suggest that it is inappropriate to apply analyses of central 



from 



human 



cratic rate of pollen introduction. Given the context, establishment of a mean 
implies nothing more than an average level of ingestion. Coprolites completely 
lacking pollen are not unknown; one such occurs among the Bighorn Cave sam- 
ples, raising questions about the implications of both pollen-negative and pollen- 
positive coprolites. The point of coprolite pollen analysis is to distinguish pollen 



mouth 



more 



means 



modern 



tions as baselines with which to compare coprolite pollen concentrations. 

In just such a support of their evaluation of the Ephedra frequencies, Reinhard 
et al. (1991:126) cited the spectra of five modern surface pollen samples taken in an 
Ephedra-dominated ecosystem near Corpus Christi. Those relative frequencies 
ranged from 8% to 45%; corresponding concentration values were not given By 
way of comparison, my own surface sampling of vegetation communities in New 
Mexico (which include Ephedra as a minor member) indicates that the time of year 
during which the sample is taken is critical. Samples taken during the pollination 
season for this wind-pollinated plant yield nearly double the calculated pollen 



same 



1991; 2723 Ephedra grains/g in June 1990; 1528 Ephedra grains/g in September 
1990). Corresponding relative frequencies for these concentration values are 12 /o 



gument 



maximum for the normal pollen rain in a v 
is present but not dominant (as proposed t 
West Texas areaV four Biehorn Cave sam 



110 



NEWS 



Vol. 13, No. 1 



more 



from well-described vegetation communities 



area of each archaeological site to establish the range of Ephedra (and other) pollen 
concentrations released into the air to be ingested by local residents. 

NAN Ranch data.— Reinhard et al. (1991) also presented and discussed pollen data 



samples from a midden 



Mexico 



previously included in discussions by Shafer et al. (1989). I calculated pollen con- 
centration values for the various pollen taxa as described earlier, using data pro- 
vided in Reinhard et al. (1991; Table 5) and Shafer et al. (1989; Table 1). I assumed 
that "trace" uniformly indicated 0.5% (1 pollen grain in a 200-grain count) of the 



« i " 



means that some of my calculations for trace per 



lated pollen concentrations are presented in Table 2. 



My 



sam 



from 



Burial Soil Sample 2), and from beneath the pelvis (here, Burial Soil Sample 3). 
This suggests that Burial Soil Sample 2 from atop the pelvis may also contain 
pollen originating from within the abdomen of the body. Reinhard et al. (1991) 
lumped all three soil samples together as control samples. These authors observed 
that pollen frequencies were dominated by Cheno/Am (goosefoot and pigweed) 
and Poaceae (grass) pollen types, and concluded that these reflected the natural 
pollen rain of the site area (Reinhard et al. 1991: 126-127), but it should be noted 
that relative frequencies for Burial Soil Sample 3 are based on an incomplete count 
(124 pollen grains; Shafer et al. 1989:22). The lower frequencies of Cheno/Am and 
Poaceae pollen types in the coprolite received no further attention. 



sam 



more 



sam 



nated sunflowers) and Zea (corn) pollen types are also more abundant in the 
coprolite, in contradiction of the pattern yielded by the relative frequencies. The 
spectrum of Burial Soil Sample 2 from atop the pelvis contains the second highest 
number of Zea pollen grains (up to twice that of the other soil samples), suggesting 



may 



must 



exposed to the natural pollen rain of the site area while midden deposits were 
accumulating. The observation that more pollen grains were introduced into a 



midden highl 
engrains may 
very unlikely 



nated from the natural pollen rain of the site area. Concentrations of Zea pollen in 

* % I "* _^ A _ A_ 



sam 



from 



grave as originally suggested by Shafer et al. (1989:27). 

The picture of pollen abundance in the burial coprolite as reflected by con 
centration values indicates that the individual consumed Cheno/Am, Poaceae 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 111 



and Zea pollen (probably along with ground chenopod and grass seeds and one or 
more ground corn products), and the pollen of low-spine Asteraceae (possibly as 
part of a tea such as made from blooming Thelesperma [Navajo tea or cota] in the 
greater Southwest to this day), and the pollen of Salix (possibly as a medicinal tea 
or a fresh food source in the early spring). The burial coprolite had been pre- 
viously interpreted as reflecting a special diet (Shafer et al. 1989), primarily on the 
basis of the lack of fiber content. Given that many post- Archaic coprolites were 
found to be fiber-free in the studies cited by Shafer et al. (1989), there would 
appear to be little in either the macrofossil or pollen portions of the NAN Ranch 
coprolite to support the suggestion that it resulted from an unusual diet. 



POLLEN EVIDENCE OF MEDICINAL USES OF PLANTS 



Reinhard et al. (1991) further concluded that presence of large frequencies of 

• a _* _ _ _ _ 



some 



(medic 



missed as "accidentally ingested" as noted earlier). This conclusion followed from 
their observations that the coprolites contained high frequencies of disaggregated 
pollen grains and lacked botanical macroremains. Because flowers usually contain 
anthers, and anthers produce pollen, the lack of pollen aggregates in the study 



floral 



fl 



fol 



plant" [page 129], and, "In the process of soaking and possibly heating vegetative 



fl 



foliage 



Medicinal 



was ascribed to the pollen data because of ethnopharmacological records for these 
taxa (page 119). 

Salix is a dioecious genus (Correll and Johnston 1979:448), with completely 
separate male and female plants; Ephedra is usually dioecious (Correll and John- 
ston 1979:80). Ingestion of unpollinated female flowers of either Salix or Ephedra 
would yield no pollen grains to a coprolite. Ingestion of foliage, bark, or stems 



female 



from male 



material. Presence of pollen implies the use of primarily male flowers and male 
plants, and secondarily the use of fertilized female flowers and pollen-dusted 
female plants. Male and female Salix plants produce precocious catkins before or 
at the beginning of leaf formation, and flowers are unlikely enough to be found in 
the field that separate keys have been devised for vegetative and floral characters 
(Correll and Johnston 1979:449-451). Ingestion of male or fertilized female flowers 
would result in the appearance of Salix pollen grains in coprolites. Taxonomic 
evidence indicates that there is a limited time frame for Salix foliage to be pro- 



accumulate 



time 



from 



proposition should be confirmed by laboratory experiments using Salix foliage 



112 NEWS and COMMENTS Vol. 13, No. 1 



gathered from male and female plants during the pollination season as well as 
later in the year. 

Why the lack of pollen aggregates which should have been present in the 
anthers of male flowers and on the fertilized female flowers? At least two possible 
solutions present themselves. Casual inspection of pollen data from my own work 
and the work of other palynologists reveals that aggregates of many pollen types 
are unusual no matter what the sample context. This is to say, pollen taxa such as 



com mo 



_ r — „„ — — } ^^». — „ 00 .^ _ 

most other pollen taxa are rarely seen as aggregates. This suggests that a lack of 
pollen aggregates is more likely a reflection of plant reproductive biology than 
human behavior. The presence of pollen aggregates, on the other hand, is cause 
for scrutiny. Lack of aggregated Ephedra pollen grains in the coprolites is a perfect 
illustration of this point, because it is a wind-pollinated taxon and pollen grains 

J 1 «.1 •• ■*-., 



minimum 



j o~ — ■ ~*— — »«- -— -- -i — 

tion from one plant to another. I can recall seeing aggregates of Ephedra pollen 



m 



in the laboratory to see whether they yield pollen aggregates. 



minute 



cedure may be especially pertinent for Salix (sonication "... separates the micro- 
scopic particles," Reinhard et al. (1991:120). It is entirely possible that this treat- 
ment disrupted pollen aggregates from male or pollinated female flowers without 



riment to extract pollen from 



microscope 



sonication would shed light on this issue. Disaggregation of pollen aggregates in 
hot tea water, as suggested by Reinhard et al. (1991:128), is unlikely. Hot water 
washes are a common part of my pollen extraction procedures and aggregates of 
other pollen types, both wind- and insect-transported, are common in the final 
pollen preparation (I have no data specifically for Salix). In the final analysis, lack 
of pollen aggregates does not in and of itself support the interpretation that 



medicinal 



Medicinal 



Moo 



Tierney 1983:70; Vines 1986:95, 104). Only twigs and bark were described for 
medicinal purposes; these were administered internally as a tea or externally as a 
wash. Castetter and Bell (1951:203) mentioned that "... a fine drink . . ." was 
formerly made of Salix flowers by the Yuma Indians on the Colorado River, but no 
medicinal effects were mentioned for this drink, nor whether male or female flow- 
ering catkins (or both) were selected for use, nor was the preparation (by infusion, 
fermentation or another method that might affect the amount of pollen present in 
he final drink) described. Lack of an ethnographic reference for a particular use 
for a particular plant does not, of course, mean that the plant never enjoyed such a 
use. Lack of an ethnographic reference, however, would suggest caution in the 
conclusion of a "new" usp in nrphict^,, 



new" use in prehistory 



some 



a me 



j "-"m"i5«-u:>iieu. ruou uses or jinn uunw—o 

catkins were lacking in the ethnographies I consulted, but Cottonwood (Papains) 



March 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



113 



liest spring wild foods in Arizona and New Mexico (Castetter 1935:43). In this 
dioecious genus, the male catkins are conspicuous but the female catkins are not. 
Elmore (1943:37-39) states that the Navajo chew Populus catkins as gum, alone or 
mixed with animal fat, and notes that the Navajo words for the flowering catkins 
of both Populus and Salix are identical. This observation suggests at least one 
native taxonomy that lumps uses of both plants together, raising the possibility 
that such nonmedicinal uses for Populus catkins as food or chewing gum could 
reasonably have extended in the past to include the catkins of closely-related Salix. 
Finally, Reinhard et al. (1991:130) suggested that Ephedra teas may have been 
used to relieve stuffy noses and colds. However, they stated earlier (page 119) that 
North American species of Ephedra do not contain ephedrine, the compound used 
to treat such symptoms and found only in Old World species of Ephedra. 



ACKNOWLEDGEMENTS 



I thank Karl J. Reinhard for scholarly discourse on the theoretical slant of this comment 
in advance of its publication; two anonymous Journal of Ethnobiology reviewers for their 
spirited comments; and Stephen A. Hall, Richard G. Holloway, and Timothy J. Seaman for 
reality checks. 



LITERATURE CITED 



WALTER 



LANDER. 1924. The rate of progress 
of food residues through the bowel. 
Journal of the American Medical Asso- 
ciation 83:576-580. 

BARKLEY, FRED A. 1934. The statistical 
theory of pollen analysis. Ecology 
13:283-289. 

BENNINGHOFF, W.S. 1962. Calculation 
of pollen and spore density in sedi- 
ments by addition of exotic pollen in 
known quantities. Pollen et Spores 
4:332-333. 

BIRKS, H.J.B. and A.D. GORDON. 1985. 
Numerical Methods in Quaternary Pol- 
len Analysis. Academic Press, London. 

CASTETTER, EDWARD F. 1935. Unculti- 
vated Native Plants used as Sources of 
Food. University of New Mexico, Bul- 



chaic Archeological Sites LA 16197, LA 
16198, and LA 16663, Bolack Exchange 
Lands, San Juan County New Mexico. 
Castetter Laboratory for Ethnobo- 
tanical Studies, Technical Report 
311:1-35. Department of Biology Uni- 
versity of New Mexico. 
DIMBLEBY, G.W. 1957. Pollen analysis of 
terrestrial soils. New Phytologist 56: 

17-28 

ELMORE, FRANCIS H. 1943. Ethno- 
botany of the Navajo. University of 
New Mexico, Bulletin 392:1-136. 
. 1976. Shrubs and Trees of 



letin 266:1-62. 



WILLIS 



Yuman Indian Agriculture: Primitive 
Subsistence on the Lower Colorado 
and Gila Rivers. University of New 
Mexico Press, Albuquerque. 

CORRELL, DONOVAN S. and MAR- 
SHALL C. JOHNSTON. 1979. Manual 
of the Vascular Plants of Texas. The 
University of Texas Press, Dallas. 

DEAN, GLENN A. 1991. Analysis of Pol- 
len and Flotation Samples from Ar- 



the Southwest Uplands. Southwest 
Parks and Monuments Association, 
Globe, Arizona. 

KELSO, GERALD K. 1976. Absolute Pol- 
len Frequencies Applied to the Interpre- 
tation of Human Activities in Northern 
Arizona. Unpublished Ph.D. Disser- 
tation, Department of Anthropology, 
University of Arizona, Tucson. 

MAHER, LOUIS J., JR. 1981. Statistics for 
microfossil concentration measure- 
ments employing samples spiked with 
marked grains. Review of Palaeobot- 
any and Palynology 32:133-191. 

MARTIN, LARRY K. 1965. Randomness of 
particle distribution in human feces 
and the resulting influence on hel- 



114 



NEWS and COMMENTS 



Vol. 13, No. 1 



minth egg counting. American Journal 
of Tropical Medicine and Hygiene 
14:747-759. 

MARTIN, PAUL S. 1963. The Last 10,000 
Years. University of Arizona Press, 
Tucson. 

MOORE, MICHAEL. 1979. Medicinal 
Plants of the Mountain West. Museum 
of New Mexico Press, Santa Fe. 

REINHARD, KARL J., DONNY L. HAM- 
ILTON, and RICHARD H. HEVLY. 
1991. Use of pollen concentration in 
paleopharmacology: Coprolite evi- 
dence of medicinal plants. Journal of 
Ethnobiology 11:117-132. 

SH AFER, HARRY J., MARIANNE MAREK, 
and KARL J. REINHARD. 1989. A 
Mimbres burial with associated colon 
remains from the NAN Ranch Ruin, 
New Mexico. Journal of Field Archae- 



TIERNEY, GAIL D. 1983. Roadside Plants 
of Northern New Mexico. The Light- 
ning Tree Press, Santa Fe. 

VINES, ROBERT A. 1986. Trees, Shrubs, 
and Woody Vines of the Southwest. 
The University of Texas Press, 
Austin. 

VON POSX LENNART. 1918. Skogastrad 



pollen 



i 



sydvenska torvmosselager- 



foldjer. Forhandlingar Skandinavika 
Naturforskeres 16, mote 1916:432-465. 
. 1967. Forest tree pollen in 



ology 16:17-30. 



south Swedish peat bog deposits (trans- 
lation by M.B. Davis and K. Faegri). 
Pollen et Spores 9:375-401. 
WILLIAMS-DEAN, GLENNA. 1978. Eth- 
nobotany and Cultural Ecology of 
Prehistoric Man in Southwest Texas. 
Unpublished Ph.D. Dissertation, De- 
partment of Botany, Texas A & M Uni- 



versity. 



RESPONSE 



THE UTILITY OF POLLEN CONCENTRATION IN COPROLITE 



Department of Anthropology, University of Nebraska 
126 Bessey Hall, Lincoln, NE 68588-0368 



COMMENTS by Karl J 



Glenna Dean presents a variety of comments on recent research into coprolite 



_ u^jvuku picuus ^\einnara et ai. iwi). borne ot her comments pro- 
vide a different perspective on the ethnobotanical significance of our finds. Other 
comments address methodological issues. The breadth of her comments prevents 
me from addressing all in a single response; consequently I will focus on those 
that are more stimulating to me. 

^ Dean comments on two papers, Reinhard et al. (1991) and Shafer et al. (1989); 

l.ttPr W *« n„KH.W ^ ^ u _, .,«_„. . ^ ^ f ^ ^^ ignQre 



Journal of Field 



.... „»„„...„ 1C6aiulllg u. concerning Keinhard et al. (1991), Dean's comments 
about plant ecology, pollen dispersion, ecological sampling, and the ethnographic 
literature were considered by us, and some of these issues were addressed in the 
original paper. Consequently I will not address them aeain althoueh I recognize 



that Dean's alternative interpretations are valid. 



comments 



""*" v " &*«-«"«•• vmue to me regard the interpretation or pouen 

concentration data. Dean's observations on this issue provide a stimulating depar- 
ture for further exploration of the pollen concentration technique. I take this op- 
portunity to place her comments within the perspective of coprolite methodology 
and to expand upon her observations. 

Although pollen analysis of coprolites has long been used for dietary recon- 
struction, it is a technique that is undergoing continuing refinement (Reinhard 
and Bryant 1992). A major goal of such analysis is separating the dietary compo- 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 115 



nent of coprolite pollen data from the nondietary. Using relative frequency data, 
several researchers have addressed this issue with some success (Bohrer 1981; 
Bryant 1974a, 1974b; Bryant and Williams-Dean 1975; Clary 1984; Scott 1979; Wil- 
liams-Dean 1978). Furthermore, the intestinal passage of pollen has been exam- 
ined experimentally by Kelso (1976) and Williams-Dean (1978) which greatly eluci- 
dated the nature of pollen transport and deposition from ingestion to defecation. 

Pollen concentration is the latest development of coprolite pollen analysis. 
The pollen concentration technique was originally devised by stratigraphic pal- 
ynologists as a means of determining absolute pollen content per unit of sediment 
and now has recognized potential in paleoethnobotanical analysis (Bryant and 
Holloway 1983; Holloway and Bryant 1986; Pearsall 1989). Although Kelso (1976) 
applied the technique in his study of modern feces, Aasen (1984) was the first 
to apply it to coprolites, followed by Sobolik (1988). Most recently, methods for 
its application to latrines have been defined by Warnock and Reinhard (1992). 
Although I had been involved in coprolite pollen analysis since the early 1980s 
(Reinhard 1985), I did not begin to apply the pollen concentration technique con- 
sistently until 1986 with an analysis of Archaic coprolites in southwest Texas 
(Reinhard et al. 1993). Since then, I have collected pollen concentration data from 
155 coprolites from Archaic and horticultural sites in the Southwest and am col- 
lecting similar data from Mesoamerican and Peruvian coprolites. 

I suspected that application of pollen concentration data to coprolites might 
provide insight into dietary behavior. In Reinhard et al. (1991) a case was pre- 
sented that pollen concentrations, when combined with relative pollen frequency 
data, help shed light on use of medicinal plants, an aspect of plant use which is 
cryptic in the archaeological record relative to dietary plant use. However, we 
were conservative in our approach and suggested that pollen concentration be 
applied as one of a battery of tests, including relative pollen frequency and statis- 
tical evaluation, to determine with greater certainty whether human behavior 
affected pollen recovered from coprolites. Multiple tests of an archaeological prob- 
lem minimize the possibility of making an interpretive error. Relative frequency 
data and pollen concentration data both reflect human use of plants, but the mag- 
nitude of pollen concentrations from coprolites provide unequivocal evidence of 
human utilization of economic taxa. We were, however, concerned about applying 
a stratigraphic technique to archaeological coprolite analysis. There are potentially 
undefined factors of human behavior and intestinal physiology that may pro- 
foundly affect pollen concentration data as has been suggested by Kelso (1976) 

and Dean (Williams-Dean 1978). 

Dean's new comments are significant contributions to the application of the 
pollen concentration technique. She suggests that pollen concentration data tell 
us more about plant use than relative frequency data. This suggests that pollen 
concentration data can be presented independently of relative frequency data. If 
correct, Dean has made a breakthrough in refining pollen analysis of coprolites. I 
test her assertion by presenting preliminary statistical evaluation of pollen values 
from coprolites (further statistical studies are planned). Dean highlights coprohte- 
specific factors that affect pollen content based on her work with modern feces. 1 
expand on this point by describing recent results of examining mummy intestinal 
contents which further point to factors of ingestion and food type that affect 



116 NEWS and COMMENTS Vol. 13, No. 1 



pollen concentration. Another issue raised by Dean relates to the utility of pollen 



consum 



m 



derived from pollen extractions of coprolites using different techniques. 



POLLEN CONCENTRATIONS VERSUS RELATIVE FREQUENCY 



Pollen data have long been presented by coprolite researchers as relative fre- 
quency expressions of a minimum of 200-grain counts. If a new technique is to be 
useful, it must provide information not available through previous approaches. 
Statistical analysis of relative pollen frequencies and pollen concentration values 
should discern whether or not pollen concentration provides different information 
than relative frequencies. 

Dean intuits that pollen concentration values provide a more complete picture 
of plant utilization than do relative frequencies. Although I agree, I wonder whether 
the two approaches provide results which are statistically different. Both are mea- 
sures of pollen abundance and as such may simply represent two different ave- 
nues to the same conclusion. For example, if relative frequency shows that a given 
pollen type makes up 90% of the pollen from a gram of coprolite and the pollen 
concentration value shows that it occurs in 1,000,000 grains per gram of coprolite, 
then both expressions indicate a high percentage and high human usage. The 
difference is the magnitude of pollen variation. Relative frequency data are nor- 
malized within 1-100% limits while concentration values typically range between 
to several million. Thus, the difference, if any, in information conveyed by the 
two techniques results from differences in magnitude of the different expressions. 

To evaluate the difference between the two expressions, I compared relative 
frequency and concentration value data for 20 coprolites, 10 from the hunter- 
gatherer site Bighorn Cave, Arizona, and 10 from the horticultural site Salmon 
Ruin, New Mexico. I selected these sites because of clear differences in the nature 
of pollen yield between them. Salmon Ruin coprolites tend to contain fewer pol- 
len types, a generally high amount of pollen per coprolite, and lessened back- 
ground pollen representation. The reverse is true for Bighorn Cave. 

The coefficient of variation was determined for each pollen taxon by site. If 
there is different information being conveyed by pollen concentration as opposed 
to relative frequency data, then one would expect that when the coefficients are 
plotted against each other, they would deviate from a linear distribution. As shown 
in Fig. 1 and Fig. 2, the two sites show variable patterns of coefficient correlation. 
Salmon Ruin coefficients show poor correlation (R 2 = 0.50); Bighorn Cave data 
show a strong correlation (R2 = 0.87). These results indicate that the two tech- 
niques do provide different information, but that the extent of difference is related 
to site. 

Why is this the case? Pollen from Bighorn Cave is largely derived from back- 
ground taxa, whereas at Salmon Ruin pollen is largely derived from dietary plants. 
It would therefore appear that human usage of plants could account for the varia- 
tion. The higher magnitude of variation exhibited by pollen concentration data is 
probably related to the higher magnitude present in the raw pollen concentration 
values. 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



117 



b 



□ 



□ 



tP 



CO 

CD 




03 






O 

CL 

CD 



o 

CO 

-Q 
< 



X 



n 



□ 



'••a 



n \ 



\ □ 



\ a 



a 



□ \ 



D 



i 



a 



D 



a 



a 



CD 



i 



* 

* 



□ 



□ 



□ 



D 



a 



D 






CO 



CO 



in 



OJ 



I/} 



CO 






03 




C 
_0 

o 

CL 








CL 



o 



o 



CO 



CO 






CM 



LO 



ID 
O 



o 



c 

— 





CD 

CL 






o 



a> 



«3 
x 

03 



o 

03 







T3 



en 





a; 

y 

3 



03 
> 



rz 

c 




c 



cu 





o 



2 P 



2 

3 






□ 



o 




D) 



C 
(0 

C 

c 



cd 

aj 



03 



2 T3 






II 



8-° 



c 
c 



"3 



03 
> 





CD 



IS 

h 

l-H 



M C 



c 



CJ 



< .a 



CD 

0> 



CD 

c 

U 
I 



u 



u- 



o 

CL 

o 

u 



C 

03 

3 

c 

Si 

3 

CL 



o 



118 



NEWS and COMMENTS 



Vol. 13, No. 1 



B± 



□ 



\ □ 



□ 



D 



□ 



□ 



□ 



CO 

CD 




CO 




CD 



O 

CO 

< 




□ 



□ 



a 



rP 



□ 



a 



1 




a 



CO 



CO 



IT) 

CM 



CO 
CD 




CX3 




c\j 



CD 

> 




DC 



J5 

CD 

DC 




E ! 

cd : 

CO 




LO 



o 



OS 



CO 



LO 



OJ 



m 



o 



c o 

> 

o 

c 

o 



T3 

CD 

> 

o 

u 

03 
X 

03 



cd 












c 



4= 



CD 

> c 

C? a; 

5- 



X5 



C 

03 

C 

o 



c 

03 



CD 

3 

CD 
CD 



03 



03 



73 

C 



m 



c 

o 



d> II 
° « 




CX 



o 

c 





c 

o 



03 



03 
> 



O 

CD 





03 

"S 

J-i 

- 

o 

o 







03 

u 



< 

c/3 



O 

U 



(N 

6 



o 

a- 

o 

c 



03 

a; 
O 



M- 



3 



o 

CD 




Summer 1993 JOURNAL OF ETHNOBIOLOGY 1 19 



Finally, to test whether there is different information conveyed on a popula- 
tion level, cluster analysis was performed on pollen concentration and relative 
frequency data from each site. With regard to Bighorn Cave, the same nine sam- 
ples sorted into one cluster while one was sorted into another for both concentra- 
tion and relative frequency data. Clustering differed for Salmon Ruin in that one 
sample was sorted out in the pollen concentration value analysis but another was 
sorted out in the relative frequency analysis. This indicates that the different tech- 
niques provide different statistical information. Further analysis of the relative 
significance between pollen frequency and pollen concentration needs to be done; 
we are continuing to evaluate this problem for the sites mentioned above. 

A difference in the information gained by applying the two techniques has 
been tentatively established by comparative analysis of a small set of coprolite 
pollen data. Now it is important to ask what factors affect concentration values. 
We should consider the influences of coprolite substrate and intestinal passage, 
for example, on pollen concentration values. A recent analysis of mummified 
intestinal contents is informative in this regard. 



POLLEN CONCENTRATION AND MUMMIES 



Dean notes that pollen concentration and relative frequency in modern feces are 

m\ & m\ M. 4 j * Mm r _ 1 



amon 



Williams-Dean 1978). Mode 



up to 32 days after consumption. Pollen is passed most abundantly two to lour 
days after consumption, then in ever-decreasing amounts thereafter. 

These studies have been critical in establishing the rate and amount of pollen 
passage through the digestive tract. My comments are not meant to diminish the 
contributions of Dean and Kelso. I am, however, concerned with directly applying 
these data to prehistoric peoples who had a very different diet with respect to fiber 
consumption, and eating habits that at times probably saturated the intestinal 
tract with pollen from a few species. Hunter-gatherer coprolites I have examined 
from Arizona, Utah, and Texas show that the majority of diet comes from a feu 
plant taxa (Reinhard 1992). This probably relates to binge eating of seasonally 
abundant foods, a phenomenon also reported for modern hunter-gatherers (Co- 
hen 1989). With respect to pollen consumption, it is possible that hinging on a 
polleniferous food may alter the pattern described from modern fecal studies. One 
might expect the intestinal tract to become saturated with specific pollen types that 
are passed for many days in large quantities. Prehistoric diets are also high in fiber 
content. High fiber content accelerates passage through the intestinal tract and 
therefore may result in defecation of pollen over shorter periods of time but in 
higher concentrations per unit measure of feces. 

Sobolik (1988) provides an example of the error in applying modern studies 
directly to coprolites. She attempted to ascertain the interval of time lapsed be- 
tween pollen consumption and defecation based on pollen concentration values. 
The basic flaw with this study is the unquestioning application of modern studies 
to a prehistoric, high fiber diet which exhibits considerable evidence of b.ng.ng. 
Sobolik presented minimal consideration of these aspects of prehistoric diet. She 



120 



NEWS and COMMENTS Vol. 13, No. 1 



also neglected Kelso's (1976) caveats regarding interpreting pollen concentration 

based on his analysis of modern feces. 

Turpin et al. (1986) demonstrated that pollen was recoverable from intestinal 
contents of mummies. To gain an understanding of pollen passage through pre- 
historic people, I have initiated study of pollen concentrations in mummified indi- 
viduals, beginning with mummified or partly mummified individuals from the 
Southwestern United States, and continuing in Peru. My goal is the recovery of 
mummified intestinal contents from different parts of the large intestine. 

Unfortunately, finding mummies with full intestines is rare; only about 30% 
of mummies contain mummified intestinal remains and of these, fewer still have 
feces in various parts of the colon. Since I have not finished the analysis of all 
Peruvian mummies, my sample size is limited to five individuals. These are none- 
theless instructive with respect to variability of pollen concentrations in mum- 
mified individuals and therefore relate to interpreting pollen concentrations of 

coprolites from latrine contexts. 

As a study area, the large intestine has several advantages over the stomach or 
small intestine. First, multiple meals can be recovered from the large intestine. 
Secondly, the amorphous, more fluid nature of chyme in the stomach and small 
intestine forms into defined, harder feces after a few hours in the large intestine. 
Thirdly, the large intestine has well defined areas that are easily identifiable in 
mummified individuals: the ascending, transverse, descending, and sigmoid por- 
tions of the colon. Coprolites can also occasionally be recovered from the rectum. 
Fourthly, although water content of the feces in the colon is inconsistent (there is 
greater water absorption on the cecum and ascending colon), it is more consistent 
than in the small intestine where water absorption is greatest. In the case of ana- 
lyzing samples, the process of natural mummification results in consistent de- 
hydration of the feces. Pollen concentration of chyme and feces may be affected by 
water content in living individuals, but this is probably not a factor in dead, desic- 
cated individuals. Thus, in the large intestine, distinct masses of concentrated, 
dehydrated fecal material can be recovered in anatomical association (reflecting 
the location and shape of distinct haustra) that are relatively consistent in water 
content. The study of these remains allows one to evaluate the passage of pollen 
through the intestine as reflected by pollen concentration values. 

Occasionally, partially mummified individuals are excavated, but anatomical 
association of the large intestine contents are not noted. Two such cases are dis- 
cussed below. In one of these, it was possible to fit the separate coprolites together 
to reconstruct their approximate association. However, it was not possible to de- 
termine which end of the series was higher in the intestinal tract. It was 
possible to sample different regions of the colon. 

One of the major factors that affects pollen concentration in feces is the nature 
of food eaten. Foods vary widely in pollen content and component digestibility 
Obviously foods with high pollen content will produce coprolites with pollen. 
However, prehistoric peoples typically had diets high in nonsoluble carbohydrate 
(fiber) in the form of vascular bundles, seed testa, epidermis, and other plant 
structures. Thus, the amount of fiber in the diet will affect the concentration of 
pollen in feces. The only incorrect assertion in Dean's review is that "many post- 
Archaic coprolites were fiber free." I have analyzed 373 coprolites from post- 



s 



till 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 121 



Archaic Southwestern sites (Reinhard 1992) and consistently find fiber to be the 
dominant component of coprolites. The nature of fiber differs between Archaic 
and post-Archaic coprolites with vascular tissue being more common in Archaic 
coprolites and seed testa more common in post-Archaic. Perhaps this is the dif- 
ference Dean is referring to, not to the general fiber content of coprolites. 

Reinhard and Hevly (1991) present the results of a study that illustrates the 
effects of different pollen and fiber concentrations on mummified feces. Four cop- 
rolites were recovered from a partially mummified individual. Two were black and 
granular and two were light brown and fibrous. One of each type was analyzed. 
We determined that the dark coprolites came from adjacent haustra and the light 
coprolites from adjacent haustra, but it was impossible to determine which pair of 
coprolites was higher in the digestive tract. 

Examination of remains from one dark, granular coprolite revealed ground 
saguaro cactus seed (Carnegiea gigantea) and fibers of mesquite pods (Ptosapis sp.). 
To quantify the remains, the hila of the ground seeds were counted. A minimal 
count of 439 saguaro seeds per gram of coprolite was obtained. In addition 11 
insect fragments, 13 mesquite pod fragments, 1 grass leaf, and 5 unidentified 
seeds were observed. In contrast the macroscopic remains of a fibrous coprolite 
were dominated by mesquite pod fragments. Two peduncles of mesquite pods 
were found among 46 mesquite pod fragments. We also identified 105 saguaro 

seeds, mesquite leaf, and a human hair. 

Palynological examination revealed a preponderance of cactus pollen in both 
granular and fibrous coprolites. The pollen concentration value of the granular 
coprolite was 500,000 grains/gm. Of these, 487,500 were consistent in morphology 
with saguaro pollen, 11,000 with Opuntia sp., and 1,500 were from nondietary, 
wind pollinated plants. From the fibrous coprolite, only 60,000 pollen grains per 
gram were recovered, 35,000 of a type similar to saguaro, 6,000 from Opuntia, and 
the remaining from nondietary, wind pollinated plants. 

I believe that the dark, granular coprolites are the residue of saguaro seed 
"cake" and the fibrous coprolite represents a meal of whole mesquite pods. The 
Opuntia sp. pollen is the residue of yet another meal, or the result of indiscriminate 
pollination of Opuntia cactus to saguaro. I believe that the saguaro pollen was 
probably ingested with the seeds since saguaro flowers are not dehiscent and 
therefore contamination of seeds with pollen while extracting the seeds and eating 

mature fruit is a strong possibility. 

This mummy, with two meals represented, reflects the effects of dietary pol- 
len and fiber content on pollen concentrations. The mesquite meal was low in 
pollen and high in insoluble fiber. Although the saguaro seed meal had a high 
fiber content in the form of seed testa, it was also high in pollen. In this case, the 
remarkable reduction in pollen concentrations of the mesquite meal (487,500 
to 35,000 saguaro-like pollen) is due at least in part to the introduction of large 
amounts of fiber. The saguaro seed content of the fibrous coprolite was about , 
that of the dark, granular coprolite, but the saguaro pollen content was reduced to 
about Vu. This shows that fiber content of a meal has a pronounced impact on the 
pollen concentration of the coprolite resulting from that meal. 

Examination of coprolites from the Dan Canyon burial (Dominguez et al. 
1992) provided an instructive case to evaluate whether pollen concentration in 



122 



NEWS and COMMENTS Vol. 13, No. 1 



feces from a single individual varied independently of fiber content. Twenty cop- 
rolites were recovered from the excavation of this partially mummified individual. 
After reconstructing the anatomical order of these coprolites, I sampled three 
different regions of the colon, probably the ascending, transverse, and descend- 
ing regions. Three coprolites (A, B, and C) were selected from these areas and 
processed. Macroscopically, all twenty coprolites were composed exclusively of 
finely ground grass testa. Thus, in this individual there was consistent dietary 
residue, and therefore consistent fiber content, between the portions of the colon. 
This makes it probable that the individual ate several meals of ground grass. In- 
deed, analysis of foods buried with the individual showed a preponderance of 
On/zopsis sp. (Indian rice grass). Archaeological reconstruction of the burial also 
suggests that the individual was buried at a time of low food diversity, probably 
limited largely to uncultivated grasses. Pollen concentrations derived from this 
burial should provide an idea of the impact of "binge" eating of a single food 



source. 



It is noteworthy that pollen concentrations varied among coprolites. For sam- 
ple A, 23,800 grains/gm were recovered, for B 49,500, and for C 19,900. The vast 
majority of pollen was derived from uncultivated grass (Poaceae). Ephedra (Mor- 
mon tea) pollen was also present as were trace amounts of Pimis (pine) and Artemi- 
sia (sage brush). Pollen aggregates were noted for Poaceae and Ephedra. The varia- 
tion of pollen concentration was due to variation in grass pollen content. I think it 
very likely that grass pollen was ingested with the ground grass seed. The occur- 



emble 



Morris 



breakage on grinding stones. 

Despite the fact that food residue was consistent, the pollen concentrations 
varied. This indicates that consistent diet over a period of time does not result in 
even distribution of pollen in the intestinal tract. The varying pollen concentra- 
tions in otherwise consistent matrices indicate that there is variation due to the 
influence of undefined factors on pollen concentrations. 

Extreme variation in pollen concentration values within a short distance in the 
intestinal tract is illustrated by the study of mummy T-10, S-241 from the site of 
Chiribaya Alta, Osmore Drainage, southern Peru. Five samples were recovered: 
ascending colon, transverse colon, splenic flexure, upper end of the descending 
colon, and middle of the descending colon (Fig. 3). Five "meals" are present in the 
colon. In the ascending colon, the coprolite macroscopic residue contains crusta- 
cean fragments and manioc tissue in a fine brown matrix composed of ground 
seed testa. The coprolite in the transverse colon contained possible starch aggre- 



matnx 



matrix of amorphus, light brown material. The upper descending colon contained 



manioc 



found in the ascending colon. In the middle of the descending colon, starch aggre- 
gates, maize hulls, and crustacean fragments are present. The sequence of foods 
was consumed in reverse order of that summarized above. 

Pollen spectra from the coprolites were dominated exclusively by Cheno/Am 
grains. Pollen concentrations of Cheno/Am varied between regions of the colon: 
9,000 in the ascending colon, none in the transverse, none in the splenic flexure, 



Summer 1993 



JOURNAL OF ETHNOBIOLOGY 



123 



ascending colon 



rectum 




descending colon 



C/) 

"D 
CD 

O 




CD 
X 





CD 



FIG. 3.— Diagram of the large intestine showing anatomical regions sampled for 
coprolites. 



8,200 in the upper end of the descending colon, and 3,400 in the middle descend- 
ing colon. Higher pollen concentrations are associated with the brown matrix 
derived from finely ground seeds. The lower concentration in the middle descend- 
ing colon is probably due to partial mixture of meals between the two descending 
colon samples. The range in pollen concentration from total absence to several 
thousand grains per gram which existed in coprolites located a few centimeters 
apart is not what one would expect based on studies of pollen in modern feces. 
Gradual voidance of pollen over several days after consumption is indicated by 

4-1 . i. ,., , F~~, ,.,.,,. t>. -irrrox TU„ ,nrlAio frnm this miimmv'S 



la 



pollen 



macroscop 



throughout the length of the intestine. Perhaps high bulk diet caused this. 



mummies 



124 



NEWS and COMMENTS Vol. 13, No. 1 



contained no pollen at all. One mummy, T-325, S-3763 from Chiribaya Alta, con- 
tained coprolites in the sigmoid and descending colon. The sigmoid colon con- 
tained maize vascular tissue, other monocot vascular tissue, boiled maize hulls, 
and spongy fiber vascular tissue. The descending colon contained guava seed 
testa, molle leaf tissue, and monocot leaf tissue. The other mummy was recovered 
from Algonodal, a site near Chiribaya Alta. Coprolites were recovered from the 
caecum, the ascending, transverse, and descending colon, and rectum. The only 
identifiable material in the caecum was monocot leaf tissue. All other coprolites 
from this mummy contained boiled maize hull, woody tissue, and maize leaf 
tissue. It is unusual for maize hulls to be found without maize pollen in coprolites 
in the Southwestern United States. The fact that five coprolites from two Peruvian 
individuals contained macroscopic maize but no pollen suggests that differences 
in maize preparation techniques between the two regions may have had an impact 
on the pollen content of consumed maize. Thus, the evidence indicates that food 
preparation technique, in addition to intestinal action, affects pollen abundance as 
reflected in pollen concentration values. 

Analysis of mummy intestinal samples highlights issues relevant to the inter- 
pretation of pollen concentration values from coprolites. Admittedly, the sample 
of mummies presented above is small. We need to build on this data base, and 
some of my current research is focused on this problem. Therefore, I present the 
following observations without generalizing them to pollen concentrations in cop- 
rolites. First, there is an interaction between the amount of insoluble fiber and 
pollen content of foods that affects concentration values. Secondly, even in consis- 
tent coprolite matrices from the same individual, pollen is not distributed equally. 
Therefore, some variation can not be explained on the basis of food texture alone. 
Thirdly, in the case of one mummy, different foods resulted in differing concentra- 
tion values in coprolites located very close to each other in the intestinal tract. 
Contrary to modern studies of fecal pollen content, this indicates that pollen can 
pass through the intestine in distinct concentrations that do not readily mix be- 
tween food residues. The most important lesson to be learned from the study of 
mummies with respect to pollen data is that one must be aware of the macroscopic 
content of coprolites in order to interpret pollen concentration values. 



POLLEN AGGREGATES 



huma 



ior, and that processing technique affects the integrity of pollen aggregates. Both 



of these issues need to be addressed. 



mportance 



ological analysis based in part on her dissertation research (1968), archaeological 



some 



as evidence of human utilization. Although I have previously used pollen aggre- 
gates as evidence of fecal origin (Reinhard et al. 1992), after considering aggregate 
data from a number of coprolites, I now believe that pollen aggregates are much 
more common from nonfecal contexts than from coorolites. Furthermore, pro- 



present in coprolites. 



seem 



Summer 1993 JOURNAL OF ETHNOBIOLOGY 125 



During the past ten years I have used very different processing techniques on 



ex a 



eggs, fungal spores, and pollen grains in the same preparation. This necessitated 
moderating the chemical treatments typically used in palynological processing. 
My coprolite processing involved rehydration, disaggregation, screening, light 
acetolysis (3-5 minutes), and brief treatment (30 seconds) in 5% potassium hydrox- 
ide. Between 1984 and 1989 my emphasis was on obtaining very clean pollen 
preparations. Processing involved rehydration, disaggregation, screening, hydro- 
chloric acid, hydrofluoric acid, zinc bromide heavy density separation, sonicatio